[go: up one dir, main page]

WO2022033456A1 - Procédé de retour de mesure d'informations d'état de canal et appareil associé - Google Patents

Procédé de retour de mesure d'informations d'état de canal et appareil associé Download PDF

Info

Publication number
WO2022033456A1
WO2022033456A1 PCT/CN2021/111700 CN2021111700W WO2022033456A1 WO 2022033456 A1 WO2022033456 A1 WO 2022033456A1 CN 2021111700 W CN2021111700 W CN 2021111700W WO 2022033456 A1 WO2022033456 A1 WO 2022033456A1
Authority
WO
WIPO (PCT)
Prior art keywords
data
parameter
model
communication device
mode
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/CN2021/111700
Other languages
English (en)
Chinese (zh)
Inventor
王婷
何高宁
冯奇
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Huawei Technologies Co Ltd
Original Assignee
Huawei Technologies Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Huawei Technologies Co Ltd filed Critical Huawei Technologies Co Ltd
Publication of WO2022033456A1 publication Critical patent/WO2022033456A1/fr
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B7/00—Radio transmission systems, i.e. using radiation field
    • H04B7/02—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas
    • H04B7/04—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas
    • H04B7/06—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station
    • H04B7/0613—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station using simultaneous transmission
    • H04B7/0615—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station using simultaneous transmission of weighted versions of same signal
    • H04B7/0619—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station using simultaneous transmission of weighted versions of same signal using feedback from receiving side
    • H04B7/0621—Feedback content
    • H04B7/0626—Channel coefficients, e.g. channel state information [CSI]
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B17/00—Monitoring; Testing
    • H04B17/30—Monitoring; Testing of propagation channels
    • H04B17/309—Measuring or estimating channel quality parameters
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B7/00—Radio transmission systems, i.e. using radiation field
    • H04B7/02—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas
    • H04B7/04—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas
    • H04B7/06—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B7/00—Radio transmission systems, i.e. using radiation field
    • H04B7/02—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas
    • H04B7/04—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas
    • H04B7/06—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station
    • H04B7/0613—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station using simultaneous transmission
    • H04B7/0615—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station using simultaneous transmission of weighted versions of same signal
    • H04B7/0619—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station using simultaneous transmission of weighted versions of same signal using feedback from receiving side
    • H04B7/0621—Feedback content
    • H04B7/0632—Channel quality parameters, e.g. channel quality indicator [CQI]

Definitions

  • the present application relates to the field of communications, and in particular, to a channel state information measurement feedback method and related apparatus.
  • network devices can allocate limited power to data streams that can be efficiently transmitted through precoding, and at the same time reduce the interference between multiple terminal devices.
  • the interference and the interference between multiple signal streams of the same terminal device are conducive to improving signal quality, realizing space division multiplexing, and improving spectrum utilization.
  • the terminal device can determine the precoding matrix based on the downlink channel measurement, and through feedback, the network device can determine the precoding for data transmission based on the precoding matrix fed back by the terminal device, thereby improving signal transmission performance.
  • the base station can obtain downlink channel information through the channel state information (CSI, Channel state information) fed back by the terminal, where the CSI is mainly a precoding matrix indicator (PMI, Precoding Matrix indicator), channel quality information (CQI, Channel quality indicator), rank One or more of the Rank indications.
  • CSI channel state information
  • PMI Precoding Matrix indicator
  • CQI Channel quality indicator
  • CSI feedback there is another option to feedback through a covariance matrix, such as a broadband or long-term statistical covariance matrix.
  • a covariance matrix such as a broadband or long-term statistical covariance matrix.
  • quantizing feedback of the covariance matrix One way is that the terminal feeds back the best orthogonal DFT basis vector corresponding to the covariance matrix.
  • Another way is to quantize the M main eigenvectors of the covariance matrix with the existing codebook.
  • the CSI feedback overhead will increase.
  • the communication performance under the CSI feedback based on the above method is not good.
  • One is that the CSI is inaccurate due to the problem of quantization feedback, and the other is that only part of the CSI information is obtained based on the above method, and the communication performance is poor.
  • Embodiments of the present invention provide a channel state information measurement feedback method and a related device, which are used to reasonably arrange channel state information feedback resources and reduce channel state information feedback overhead.
  • the mode of the first model is used to determine the first model, the first model is formed based on structural data and parameter data, the first model is a channel learning model, and the channel learning model can be configured according to the parameters.
  • the channel state information is processed to reduce the amount of data of the channel state information, and the channel state information reference signal CSI-RS is obtained.
  • the CSI-RS is used to determine the first channel state information CSI, and the first model and the first CSI are used to determine the first Two CSI, the data amount of the second CSI is smaller than the data amount of the first CSI, and the second CSI is output.
  • the first communication device can determine the mode of the first model according to the actual communication situation, and the methods for determining the structural data and parameter data of the first model are different between different modes, which improves the flexibility of the solution and improves the flexibility of the solution.
  • the methods for determining the structural data and parameter data of the first model are different between different modes, which improves the flexibility of the solution and improves the flexibility of the solution.
  • the first communication device feeds back the second CSI, which can reduce the overhead of the CSI feedback compared to the first CSI.
  • the second communication device may determine the first CSI based on the second CSI fed back by the first communication device and the first model, obtain complete channel information, and improve communication performance.
  • the embodiment of the present invention also provides a first implementation manner based on the first aspect:
  • the mode of the first model is determined according to a first parameter, where the first parameter includes one or more of channel transmission scenarios, channel transmission uplink and downlink resources, and device capabilities.
  • the first communication device can predefine the mode of the channel learning model according to the first parameter, and the first parameter includes the channel transmission scenario, the channel transmission uplink and downlink resources, and the device capability of the first communication device. It interacts with the signaling of the second communication device, and while reducing the signaling interaction, the mode of the channel learning model is determined according to the actual communication situation, and the channel state information is fed back.
  • the first communication device may determine the mode of the channel learning model according to the first parameter. That is, different modes of the channel learning model can be determined according to different values or different situations of the first parameter.
  • the determination of the mode of the channel learning model takes into account the conditions of different first parameters, which can be more adaptable to different transmission scenarios, uplink and downlink resources, and device capabilities, meet the needs of different situations, improve the accuracy of determining the first model, and improve communication. performance.
  • the embodiments of the present invention further provide a second implementation manner of the first aspect:
  • the first communication device sends mode request information to the second communication device, so that the second communication device determines the channel learning model mode, and the mode request information may include one or more modes of the channel learning model.
  • the information is used for reference by the channel state feedback device, so that the second communication device can determine the channel learning model mode according to its actual communication situation and also consider the actual communication situation where the first communication device is located, which improves the flexibility of the scheme. At the same time, the rationality is further improved.
  • the embodiment of the present invention further provides a third implementation manner of the first aspect:
  • mode indication information where the mode indication information is used to indicate the mode of the first model.
  • the first communication device can determine the channel learning model mode according to the mode indication information of the second communication device, and has low requirements on the device capability of the first communication device, and only needs to receive the indication information of the second communication device.
  • the mode of the channel learning model can be determined, which improves the flexibility of the scheme.
  • the embodiment of the present invention further provides a fourth implementation of the first aspect:
  • the first communication device can obtain the structure of the channel learning model from the second communication device. class data and parameter class data, so that the first communication device builds a channel learning model according to the acquired structural class data and parameter class data, so as to determine the channel state information after reducing the amount of data.
  • the second communication device may determine structural class data and parameter class data of the channel learning model.
  • the second communication device may determine the channel learning model based on the previous channel information, for example, the first communication device for a cell or a certain area may use the same channel learning model, which reduces the complexity of determining the channel learning model.
  • the embodiment of the present invention further provides a fifth implementation of the first aspect:
  • the structure class data and parameter class data of the first model are output.
  • the first communication device when the mode of determining the channel learning model is that the first communication device determines the structural data and parameter data of the channel learning model, the first communication device can output the determined structural data and parameter data, to be acquired by the second communication device, so that the second communication device builds a channel learning model according to the structural data and parameter data, and processes the received channel state information according to the channel learning model.
  • the first communication device determines the structural data and parameter data of the channel learning model according to the actual communication situation, which is more suitable for scenarios where the channel situation is complex and the first communication device has strong device capabilities.
  • the complexity of training the second communication device to determine the channel learning model is reduced, and the energy saving of the second communication device is realized.
  • the first communication device determines the structural data and the parameter data
  • the embodiment of the present invention further provides a sixth implementation of the first aspect:
  • Acquire structural class data of the first model determine parameter class data of the first model according to the structural class data, and output the parameter class data.
  • the mode for determining the channel learning model is that the second communication device determines the structure class data and the first communication device determines the parameter class data
  • the first communication device learns the model structure class according to the channel determined by the second communication device.
  • the data further determines the parameter class data of the channel learning model.
  • the second communication device determines the structure class data, which reduces the requirement on the capability of the first communication device and increases the implementability of the solution.
  • the second communication device can determine the structural data according to the previous channel information, the second communication device can grasp the complexity of the channel learning model, reduce the complexity of the first communication device training to determine the parameter data, and avoid the first communication device.
  • Communication devices need to train different channel learning models for performance under structured data multiple times.
  • the first communication device trains the matched parameter class data based on the structural class data determined by the second communication device, which can reduce the complexity of the first communication device training and determine the channel learning model, and realize energy saving of the first communication device.
  • the parameter type data determined by the first communication device may be that the first communication device trains the channel learning model according to the channel environment, so that the parameter type data more matches the current channel environment, improves the accuracy of the channel learning model, and improves the communication performance.
  • the embodiment of the present invention further provides a seventh implementation of the first aspect:
  • the structural class data of the first model is output, and the parameter class data determined according to the structural class data is obtained.
  • the second communication device learns the model structure class according to the channel determined by the first communication device.
  • the data further determines the parameter data of the channel learning model, which reduces the amount of data that the second communication device needs to process for constructing the channel learning model in the actual solution implementation process, which is more suitable for the situation where the first communication device has strong capabilities.
  • the first communication device can determine the structural data of the channel learning model based on the capabilities of the first communication device, the first communication device can grasp the complexity of the channel learning model, and reduce the complexity of the second communication device in determining the parameter data , to achieve energy saving of the second communication device.
  • the parameter-type data determined by the second communication device may be parameter-type data determined by the second communication device according to the previous channel information of the channel learning model, for example, the first communication device in a cell or a certain area may use the same
  • the channel learning model reduces the complexity of determining the channel learning model.
  • the embodiment of the present invention further provides the eighth implementation manner of the first aspect:
  • Output structural data obtain first feedback configuration information, the first feedback configuration information has a corresponding relationship with structural data, determine parameter data according to the first feedback configuration information and structural data, and output the parameter data.
  • the first communication device feeds back the structural data and the parameter data to the second communication device.
  • Hierarchical feedback can be performed, and after the structural data is fed back to the second communication device, the first feedback configuration information of the second communication device corresponding to the structural data can be obtained, so as to achieve reasonable allocation of feedback resources and avoid resource waste.
  • the parameter data can be determined according to the first feedback configuration information and the structure data.
  • the parameter data can be further determined according to the first feedback of the second communication device.
  • a feedback configuration information determines more reasonable parameter class data.
  • the embodiment of the present invention further provides the ninth implementation manner of the first aspect:
  • the first sub-data of the parameter-type data is output, the first sub-data is included in the parameter-type data, and the first sub-data is determined according to the first feedback configuration information and the structure-type data.
  • the second communication device can construct a channel learning model in the case of receiving part of the parameter data, which reduces the impact on the channel when the first communication device feeds back the parameter data.
  • pressure which reduces the occupation of channel resources when feeding back CSI.
  • the embodiment of the present invention further provides the tenth implementation manner of the first aspect:
  • the first model is trained according to the first feedback configuration information and the structural data, and the parameter data of the first model is determined.
  • the first communication device can retrain the channel learning model according to the first feedback configuration information of the second communication device, thereby determining parameter data that is more suitable for the current channel environment. , which improves the accuracy of the parameter data determined by the first communication device.
  • the embodiment of the present invention further provides the eleventh implementation of the first aspect:
  • the second sub-data determines the third sub-data of the parameter-type data, the third sub-data is included in the parameter-type data, and the third sub-data is output.
  • the first communication device can output the parameter data to the second communication device in a hierarchical manner.
  • the second sub-data of the data and the second feedback configuration information generated after the second communication device acquires the second sub-data determines the third sub-data of the parameter-type data, which can reduce the data volume of the third sub-data and improve the third sub-data.
  • the accuracy of the data improves the flexibility and accuracy of the solution implementation.
  • the embodiment of the present invention further provides the twelfth implementation of the first aspect:
  • the parameter class data is output according to the time characteristic of the parameter class data.
  • the parameter output can be performed according to the optimal feedback cycle of each parameter data, which reduces the Channel resources occupied when feeding back CSI.
  • the embodiment of the present invention further provides the thirteenth implementation manner of the first aspect:
  • Acquire structural data obtain first feedback configuration information, where the first feedback configuration information has a corresponding relationship with structural data, and obtain parameter data, where the parameter data is determined according to the first feedback configuration information and structural data.
  • the first communication device can obtain the structural data and parameter data output by the second communication device in stages, wherein the parameter data is determined by the structural data and the first feedback configuration relationship, which avoids outputting the structure at one time.
  • the class data and the parameter class data cause greater pressure on the channel, more reasonable parameter class data is further determined according to the first feedback configuration information of the second communication device.
  • the embodiment of the present invention further provides the fourteenth implementation manner of the first aspect:
  • the first sub-data of the parameter-type data is acquired, the first sub-data is included in the parameter-type data, and the first sub-data is determined according to the first feedback configuration information and the structure-type data.
  • the first communication device can construct a channel learning model in the case of receiving part of the parametric data, which reduces the impact on the channel when the second communication device feeds back the parametric data.
  • pressure which reduces the occupation of channel resources when feeding back CSI.
  • the embodiment of the present invention further provides the fifteenth implementation of the first aspect :
  • the second sub-data of the parameter-type data is included in the parameter-type data
  • acquire the second feedback configuration information has a corresponding relationship with the second sub-data of the parameter-type data
  • acquire the parameter-type data The third sub-data of the data, the third sub-data is determined according to the second feedback configuration information and the second sub-data, and the third sub-data is included in the parameter type data.
  • the second communication device can output the parameter class data to the first communication device in a hierarchical manner.
  • the second sub-data of the data and the second feedback configuration information generated after the second communication device acquires the second sub-data determines the third sub-data of the parameter-type data, which can reduce the data volume of the third sub-data and improve the third sub-data.
  • the accuracy of the data improves the flexibility and accuracy of the solution implementation.
  • the embodiment of the present invention further provides the sixteenth implementation manner of the first aspect :
  • the parameter acquisition can be carried out according to the optimal feedback cycle of each parameter data, which reduces the number of parameters.
  • Channel resources occupied when feeding back CSI are different.
  • any of the eighth, ninth, tenth, eleventh, twelfth, thirteenth, fourteenth, fifteenth, and sixteenth types based on the first aspect also provides a seventeenth implementation manner of the first aspect:
  • At least one item of the second type of data resource of the first model, the number of bits, and the feedback information is determined according to the first feedback configuration information and the structure type data.
  • the first communication device can determine the configuration information corresponding to the parameter data according to the first feedback configuration information determined by the second communication device, and the configuration information can be determined according to different channel states.
  • One or more of these options improve the flexibility and applicability of this solution.
  • a second aspect of the present application provides a channel state information measurement feedback method, including:
  • the mode of the first model is used to determine the first model, the first model is formed based on the structural data and the parameter data, outputs the channel state information reference signal CSI-RS, and obtains the second channel state information CSI, the first CSI is determined according to the first model and the second CSI, and the data amount of the second CSI is smaller than the data amount of the first CSI.
  • the second communication device can determine the mode of the first model according to the actual communication situation, and the methods for determining the structural data and parameter data of the first model are different between different modes, which improves the flexibility of the solution and improves the efficiency of the solution.
  • the methods for determining the structural data and parameter data of the first model are different between different modes, which improves the flexibility of the solution and improves the efficiency of the solution.
  • the second communication device acquires the second CSI, which can reduce the overhead of CSI feedback compared to the first CSI. Further, the second communication device may determine the first CSI based on the second CSI fed back by the first communication device and the first model, obtain complete channel information, and improve communication performance.
  • the embodiments of the present invention also provide a first implementation manner of the second aspect:
  • the mode of the first model is determined according to a first parameter, where the first parameter includes one or more of channel transmission scenarios, channel transmission uplink and downlink resources, and device capabilities.
  • the second communication device can predefine the mode of the channel learning model according to the first parameter, and the first parameter includes the channel transmission scenario, the channel transmission uplink and downlink resources, and the device capability of the first communication device. It interacts with the signaling of the second communication device, and while reducing the signaling interaction, the mode of the channel learning model is determined according to the actual communication situation, and the channel state information is fed back.
  • the second communication device may determine the mode of the channel learning model according to the first parameter. That is, different modes of the channel learning model can be determined according to different values of the first parameter or different situations.
  • the determination of the mode of the channel learning model takes into account the situation of different first parameters, which can be more adaptable to different transmission scenarios, uplink and downlink resources, and device capabilities, meet the needs of different situations, improve the accuracy of determining the first model, and improve communication. performance.
  • the embodiments of the present invention also provide a second implementation manner of the second aspect:
  • mode request information where the mode request information is used to request to determine the mode of the first model.
  • the first communication device sends mode request information to the second communication device, so that the second communication device determines the channel learning model mode, and the mode request information may include one or more modes of the channel learning model.
  • the information is used for reference by the channel state feedback device, so that the second communication device can determine the channel learning model mode according to its actual communication situation and also consider the actual communication situation where the first communication device is located, which improves the flexibility of the scheme. At the same time, the rationality is further improved.
  • the embodiment of the present invention further provides a third implementation manner of the second aspect:
  • Output mode indication information where the mode indication information is used to indicate the mode of the first model.
  • the mode indication information of the second communication device can determine the mode of the channel learning model, and the device capability requirements of the first communication device are relatively low, and the first communication device only needs to receive the indication information of the second communication device. Determining the mode of the channel learning model improves the flexibility of the scheme.
  • the embodiment of the present invention further provides a fourth implementation of the second aspect:
  • the structure class data and parameter class data of the first model are output.
  • the second communication device can output the structure class of the channel learning model to the first communication device. data and parameter data, so that the first communication device builds a channel learning model according to the acquired structural data and parameter data, so as to determine the channel state information after reducing the amount of data.
  • the second communication device may determine structural class data and parameter class data of the channel learning model.
  • the second communication device may determine the channel learning model based on the previous channel information, for example, the first communication device in a cell or a certain area may use the same channel learning model, which reduces the complexity of determining the channel learning model.
  • the embodiment of the present invention further provides a fifth implementation of the second aspect:
  • the second communication device can obtain the structure of the channel learning model from the first communication device. class data and parameter class data, so that the second communication device builds a channel learning model according to the acquired structure class data and parameter class data, so as to determine the channel state information after restoring the data amount.
  • the complexity of training the second communication device to determine the channel learning model is reduced, and the energy saving of the second communication device is realized.
  • the first communication device determines the structural data and the parameter data
  • the embodiment of the present invention further provides a sixth implementation of the second aspect:
  • the structural class data of the first model is output, and the parameter class data is obtained, and the parameter class data is determined according to the structural class data.
  • the mode for determining the channel learning model is that the second communication device determines the structure class data and the first communication device determines the parameter class data
  • the first communication device learns the model structure class according to the channel determined by the second communication device.
  • the data further determines the parameter class data of the channel learning model.
  • the second communication device determines the structure class data, which reduces the requirement on the capability of the first communication device and increases the implementability of the solution.
  • the second communication device can determine the structural data according to the previous channel information, the second communication device can grasp the complexity of the channel learning model, reduce the complexity of the first communication device training to determine the parameter data, and avoid the first communication device.
  • Communication devices need to train different channel learning models for performance under structured data multiple times.
  • the first communication device trains the matched parameter class data based on the structural class data determined by the second communication device, which can reduce the complexity of the first communication device training and determine the channel learning model, and realize energy saving of the first communication device.
  • the parameter type data determined by the first communication device may be that the first communication device trains the channel learning model according to the channel environment, so that the parameter type data more matches the current channel environment, improves the accuracy of the channel learning model, and improves the communication performance.
  • the embodiment of the present invention also provides a seventh implementation of the second aspect:
  • Acquire structural class data of the first model determine parameter class data of the first model according to the structural class data, and output the parameter class data.
  • the second communication device learns the model structure class according to the channel determined by the first communication device.
  • the data further determines the parameter data of the channel learning model, which reduces the amount of data that the second communication device needs to process for constructing the channel learning model in the actual solution implementation process, which is more suitable for the situation where the first communication device has strong capabilities.
  • the first communication device can determine the structural data of the channel learning model based on the capabilities of the first communication device, the first communication device can grasp the complexity of the channel learning model, and reduce the complexity of the second communication device in determining the parameter data , to achieve energy saving of the second communication device.
  • the parameter-type data determined by the second communication device may be parameter-type data determined by the second communication device according to the previous channel information of the channel learning model, for example, the first communication device in a cell or a certain area may use the same
  • the channel learning model reduces the complexity of determining the channel learning model.
  • the embodiment of the present invention further provides an eighth implementation manner of the second aspect:
  • the structural data is acquired, the first feedback configuration information is determined, the first feedback configuration information has a corresponding relationship with the structural data, and the parameter data is obtained, and the parameter data is determined according to the first feedback configuration information and the structural data.
  • the first communication device feeds back the structural data and the parameter data to the second communication device.
  • Hierarchical feedback can be performed, and the second communication device correspondingly determines the first feedback configuration information according to the structural data, so as to achieve reasonable allocation of feedback resources and avoid resource waste, and at the same time, the first communication device can be based on the first feedback configuration information and The structural data determines the parameter data.
  • more reasonable parameter data is further determined according to the first feedback configuration information.
  • the embodiment of the present invention further provides a ninth implementation manner of the second aspect:
  • the first sub-data of the parameter-type data is acquired, the first sub-data is included in the parameter-type data, and the first sub-data is determined according to the first feedback configuration information and the structure-type data.
  • the second communication device can construct a channel learning model in the case of receiving part of the parameter data, which reduces the impact on the channel when the first communication device feeds back the parameter data.
  • pressure which reduces the occupation of channel resources when feeding back CSI.
  • the embodiment of the present invention further provides the tenth implementation of the second aspect:
  • the second sub-data of the parameter-type data is included in the parameter-type data
  • determine the second feedback configuration information has a corresponding relationship with the second sub-data of the parameter-type data
  • obtain the parameter-type data The third sub-data of the data, the third sub-data is determined according to the second feedback configuration information and the second sub-data, and the third sub-data is included in the parameter type data.
  • the first communication device can output the parameter data to the second communication device in a hierarchical manner.
  • the second sub-data of the data and the second feedback configuration information generated after the second communication device acquires the second sub-data determines the third sub-data of the parameter-type data, which can reduce the data volume of the third sub-data and improve the third sub-data at the same time.
  • the accuracy of the data improves the flexibility and accuracy of the solution implementation.
  • the embodiment of the present invention further provides the eleventh implementation of the second aspect:
  • the parameter data can contain multiple specific parameter data, and the optimal feedback cycle corresponding to each parameter data is different, the parameters can be obtained according to the optimal feedback cycle of each parameter data, reducing the need for Channel resources occupied during CSI feedback.
  • the embodiment of the present invention further provides the twelfth implementation manner of the second aspect:
  • Output structural data determine first feedback configuration information, where the first feedback configuration information has a corresponding relationship with structural data, determine parameter data according to the first feedback configuration information and structural data, and output the parameter data.
  • the first communication device can obtain the structural data and parameter data output by the second communication device in stages, wherein the parameter data is determined by the structural data and the first feedback configuration relationship, which avoids outputting the structure at one time.
  • the class data and the parameter class data cause greater pressure on the channel, more reasonable parameter class data is further determined according to the first feedback configuration information of the second communication device.
  • an embodiment of the present invention further provides a thirteenth implementation manner of the second aspect:
  • the first sub-data of the parameter-type data is output, the first sub-data is included in the parameter-type data, and the first sub-data is determined according to the first feedback configuration information and the structure-type data.
  • the parameter data of the output part of the second communication device can enable the first communication device to construct a channel learning model, which reduces the need for the second communication device to feedback parameter data.
  • the pressure of the channel reduces the occupation of channel resources when feeding back CSI.
  • the embodiment of the present invention further provides the fourteenth implementation manner of the second aspect:
  • the first model is trained according to the first feedback configuration information and the structural data, and the parameter data of the first model is determined.
  • the second communication device can retrain the channel learning model according to the first feedback configuration information, thereby determining parameter data more suitable for the current channel environment, which improves the second communication device. Accuracy of parametric data as determined by communications equipment.
  • the embodiment of the present invention further provides the fifteenth aspect of the second aspect Implementation:
  • the second communication device can output the parameter class data to the first communication device in a hierarchical manner.
  • the second sub-data of the data and the second feedback configuration information generated after the second communication device acquires the second sub-data determines the third sub-data of the parameter-type data, which can reduce the data volume of the third sub-data and improve the third sub-data.
  • the accuracy of the data improves the flexibility and accuracy of the solution implementation.
  • the embodiment of the present invention further provides the sixteenth aspect of the second aspect Implementation:
  • the parameter class data is output according to the time characteristic of the parameter class data.
  • the parameter output can be performed according to the optimal feedback cycle of each parameter data, which reduces the Channel resources occupied when feeding back CSI.
  • any of the eighth, ninth, tenth, eleventh, twelfth, thirteenth, fourteenth, fifteenth, and sixteenth types based on the second aspect also provides a seventeenth embodiment of the second aspect:
  • At least one item of the resource, the number of bits, and the feedback information corresponding to the parameter type data of the first model is determined.
  • the first communication device can determine the configuration information corresponding to the parameter data according to the first feedback configuration information determined by the second communication device, and the configuration information can be determined according to different channel states.
  • One or more of these options improve the flexibility and applicability of this solution.
  • a third aspect of the embodiments of the present invention provides a first communication device, including a first logic circuit and a first communication interface:
  • a first logic circuit configured to determine a mode of the first model, and the mode of the first model is used to determine a first model, the first model is formed based on structural data and parameter data;
  • a first communication interface used to obtain a channel state information reference signal CSI-RS, where the CSI-RS is used to determine the first channel state information CSI;
  • a first logic circuit further configured to determine a second CSI according to the first model and the first CSI, where the data volume of the second CSI is smaller than the data volume of the first CSI;
  • the first communication interface is also used for outputting the second CSI.
  • the first communication device can determine the mode of the first model according to the actual communication situation, and the methods for determining the structural data and parameter data of the first model are different between different modes, which improves the flexibility of the solution and improves the flexibility of the solution.
  • the methods for determining the structural data and parameter data of the first model are different between different modes, which improves the flexibility of the solution and improves the flexibility of the solution.
  • the embodiment of the present invention also provides a first implementation manner based on the third aspect:
  • the first logic circuit is specifically configured to determine the mode of the first model according to a first parameter, where the first parameter includes one or more of a channel transmission scenario, channel transmission uplink and downlink resources, and device hardware parameters.
  • the first communication device can predefine the mode of the channel learning model according to the first parameter, and the first parameter includes the channel transmission scenario, the channel transmission uplink and downlink resources, and the device capability of the first communication device. It interacts with the signaling of the second communication device, and while reducing the signaling interaction, the mode of the channel learning model is determined according to the actual communication situation, and the channel state information is fed back.
  • the embodiment of the present invention also provides a second implementation manner based on the third aspect:
  • the first communication interface is specifically configured to output mode request information, where the mode request information is used to request to determine the mode of the first model.
  • the first communication device sends mode request information to the second communication device, so that the second communication device determines the channel learning model mode, and the mode request information may include one or more modes of the channel learning model.
  • the information is used for reference by the channel state feedback device, so that the second communication device can determine the channel learning model mode according to its actual communication situation and also consider the actual communication situation where the first communication device is located, which improves the flexibility of the scheme. At the same time, the rationality is further improved.
  • the first communication device may determine the mode of the channel learning model according to the first parameter. That is, different modes of the channel learning model can be determined according to different values or different situations of the first parameter.
  • the determination of the mode of the channel learning model takes into account the conditions of different first parameters, which can be more adaptable to different transmission scenarios, uplink and downlink resources, and device capabilities, meet the needs of different situations, improve the accuracy of determining the first model, and improve communication. performance.
  • the embodiment of the present invention further provides a third implementation manner of the third aspect:
  • the first communication interface is further configured to acquire mode indication information, where the mode indication information is used to indicate the mode of the first model.
  • the first communication device can determine the channel learning model mode according to the mode indication information of the second communication device, and has low requirements on the device capability of the first communication device, and only needs to receive the indication information of the second communication device.
  • the mode of the channel learning model can be determined, which improves the flexibility of the scheme.
  • the embodiment of the present invention further provides a fourth implementation of the third aspect:
  • the first communication interface is further configured to acquire structural data and parameter data of the first model.
  • the first communication device can obtain the structural data and parameter data of the channel learning model from the second communication device. data, so that the first communication device builds a channel learning model according to the acquired structural data and parameter data, so as to determine the channel state information after reducing the amount of data.
  • the second communication device may determine structural class data and parameter class data of the channel learning model.
  • the second communication device may determine the channel learning model based on the previous channel information, for example, the first communication device in a cell or a certain area may use the same channel learning model, which reduces the complexity of determining the channel learning model.
  • the embodiment of the present invention further provides a fifth implementation of the third aspect:
  • the first communication interface is also used to output structural data and parameter data of the first model.
  • the first communication device when the mode of determining the channel learning model is that the first communication device determines the structural data and parameter data of the channel learning model, the first communication device can output the determined structural data and parameter data, for the second communication device to acquire, so as to construct a channel learning model according to the structural data and parameter data, and process the received channel state information according to the channel learning model.
  • the first communication device determines the structural data and parameter data of the channel learning model according to the actual communication situation, which is more suitable for scenarios where the channel situation is complex and the first communication device has strong device capabilities.
  • the complexity of training the second communication device to determine the channel learning model is reduced, and the energy saving of the second communication device is realized.
  • the first communication device determines the structural data and the parameter data
  • the embodiment of the present invention further provides a sixth implementation of the third aspect:
  • a first communication interface used to obtain structural class data of the first model
  • the first logic circuit is further configured to determine the parameter class data of the first model according to the structure class data;
  • the first communication interface is also used for outputting parameter data.
  • the mode for determining the channel learning model is that the second communication device determines the structure class data and the first communication device determines the parameter class data
  • the first communication device learns the model structure class according to the channel determined by the second communication device.
  • the data further determines the parameter class data of the channel learning model.
  • the second communication device determines the structure class data, which reduces the requirement on the capability of the first communication device and increases the implementability of the solution.
  • the second communication device can determine the structural data according to the previous channel information, the second communication device can grasp the complexity of the channel learning model, reduce the complexity of the first communication device training to determine the parameter data, and avoid the first communication device.
  • Communication devices need to train different channel learning models for performance under structured data multiple times.
  • the first communication device trains the matched parameter class data based on the structural class data determined by the second communication device, which can reduce the complexity of the first communication device training and determine the channel learning model, and realize energy saving of the first communication device.
  • the parameter type data determined by the first communication device may be that the first communication device trains the channel learning model according to the channel environment, so that the parameter type data more matches the current channel environment, improves the accuracy of the channel learning model, and improves the communication performance.
  • the embodiment of the present invention further provides a seventh implementation of the third aspect:
  • a first communication interface for outputting structural data of the first model
  • the first communication interface is further configured to acquire parameter class data, where the parameter class data is determined according to the structural class data.
  • the second communication device learns the model structure class according to the channel determined by the first communication device.
  • the data further determines the parameter data of the channel learning model, which reduces the amount of data that the second communication device needs to process for constructing the channel learning model in the actual solution implementation process, which is more suitable for the situation where the first communication device has strong capabilities.
  • the first communication device can determine the structural data of the channel learning model based on the capabilities of the first communication device, the first communication device can grasp the complexity of the channel learning model, and reduce the complexity of the second communication device in determining the parameter data , to achieve energy saving of the second communication device.
  • the parameter-type data determined by the second communication device may be parameter-type data determined by the second communication device according to the previous channel information of the channel learning model, for example, the first communication device in a cell or a certain area may use the same
  • the channel learning model reduces the complexity of determining the channel learning model.
  • the embodiment of the present invention further provides an eighth implementation manner of the third aspect:
  • a first communication interface specifically used for outputting structural data
  • a first logic circuit specifically configured to acquire first feedback configuration information, where the first feedback configuration information has a corresponding relationship with structural data
  • a logic circuit further configured to determine parameter type data according to the first feedback configuration information and structure type data
  • the first communication interface is also used for outputting parameter data.
  • the first communication device feeds back the structural data and the parameter data to the second communication device.
  • Hierarchical feedback may be performed, and after the structural data is fed back to the second communication device, the first feedback configuration information of the structural data from the second communication device is obtained, and the parameter data is determined according to the first feedback configuration information and the structural data, On the basis of avoiding the large pressure on the channel caused by the one-time output of structural data and parameter data, more reasonable parameter data is further determined according to the first feedback configuration information of the second communication device.
  • the embodiment of the present invention further provides the ninth implementation manner of the third aspect:
  • the first communication interface is specifically used to output the first sub-data of the parameter-type data.
  • the first sub-data is included in the parameter-type data, and the first sub-data is determined according to the first feedback configuration information and the structure-type data.
  • the second communication device can construct a channel learning model in the case of receiving part of the parameter data, which reduces the impact on the channel when the first communication device feeds back the parameter data.
  • pressure which reduces the occupation of channel resources when feeding back CSI.
  • the embodiment of the present invention further provides the tenth implementation manner of the third aspect:
  • the first logic circuit is specifically configured to train the first model according to the first feedback configuration information and the structural data, and determine the parameter data of the first model.
  • the first communication device can retrain the channel learning model according to the first feedback configuration information of the second communication device, thereby determining parameter data that is more suitable for the current channel environment. , which improves the accuracy of the parameter data determined by the first communication device.
  • the embodiment of the present invention further provides an eleventh implementation of the third aspect:
  • a first communication interface for outputting the second sub-data of the parameter-type data, and the second sub-data is included in the parameter-type data
  • the first communication interface is further configured to acquire second feedback configuration information, where the second feedback configuration information has a corresponding relationship with the second sub-data of the parameter data;
  • a first logic circuit configured to determine the third sub-data of the parameter-type data according to the second feedback configuration information and the second sub-data;
  • the first communication interface is also used for outputting third sub-data.
  • the first communication device can output the parameter data to the second communication device in a hierarchical manner.
  • the second sub-data of the data and the second feedback configuration information generated after the second communication device acquires the second sub-data determines the third sub-data of the parameter-type data, which can reduce the data volume of the third sub-data and improve the third sub-data.
  • the accuracy of the data improves the flexibility and accuracy of the solution implementation.
  • the embodiment of the present invention also provides the twelfth implementation of the third aspect:
  • the first communication interface is used for outputting the parameter data according to the time characteristic of the parameter data.
  • the parameter output can be performed according to the optimal feedback cycle of each parameter data, which reduces the Channel resources occupied when feeding back CSI.
  • the embodiment of the present invention also provides the thirteenth implementation manner of the third aspect:
  • a first communication interface used to obtain structural data
  • the first communication interface is further configured to acquire first feedback configuration information, where the first feedback configuration information has a corresponding relationship with structural data;
  • the first communication interface is further configured to acquire parameter type data, where the parameter type data is determined according to the first feedback configuration information and the structure type data.
  • the first communication device can obtain the structural data and parameter data output by the second communication device in stages, wherein the parameter data is determined by the structural data and the first feedback configuration relationship, which avoids outputting the structure at one time.
  • the class data and the parameter class data cause greater pressure on the channel, more reasonable parameter class data is further determined according to the first feedback configuration information of the second communication device.
  • the first communication device can construct a channel learning model in the case of receiving part of the parametric data, which reduces the impact on the channel when the second communication device feeds back the parametric data.
  • pressure which reduces the occupation of channel resources when feeding back CSI.
  • the embodiment of the present invention further provides the fourteenth implementation manner of the third aspect:
  • the first communication interface is used to obtain the first sub-data of the parameter-type data, the first sub-data is included in the parameter-type data, and the first sub-data is determined according to the first feedback configuration information and the structure-type data.
  • the embodiment of the present invention further provides the fifteenth implementation of the third aspect :
  • a first communication interface for acquiring second sub-data of the parameter-type data, where the second sub-data is included in the parameter-type data
  • the first communication interface is further configured to acquire second feedback configuration information, where the second feedback configuration information has a corresponding relationship with the second sub-data of the parameter data;
  • the first communication interface is further configured to acquire third sub-data of the parameter-type data, where the third sub-data is determined according to the second feedback configuration information and the second sub-data, and the third sub-data is included in the parameter-type data.
  • the second communication device can output the parameter class data to the first communication device in a hierarchical manner.
  • the second sub-data of the data and the second feedback configuration information generated after the second communication device acquires the second sub-data determines the third sub-data of the parameter-type data, which can reduce the data volume of the third sub-data and improve the third sub-data.
  • the accuracy of the data improves the flexibility and accuracy of the solution implementation.
  • the embodiment of the present invention further provides the sixteenth implementation manner of the third aspect :
  • the first communication interface is used for acquiring parameter data according to the time characteristic of the parameter data.
  • the parameter acquisition can be carried out according to the optimal feedback cycle of each parameter data, which reduces the number of parameters.
  • Channel resources occupied when feeding back CSI are different.
  • the embodiment of the present invention also provides a seventeenth implementation of the third aspect:
  • the first logic circuit is configured to determine at least one item of the second type of data resource, the number of bits, and the feedback information of the first model according to the first feedback configuration information and the structure type data.
  • the first communication device can determine the configuration information corresponding to the parameter data according to the first feedback configuration information determined by the second communication device, and the configuration information can be determined according to different channel states.
  • One or more of these options improve the flexibility and applicability of this solution.
  • a fourth aspect of the embodiments of the present invention provides a second communication device, including a second logic circuit and a second communication interface:
  • the second logic circuit is used to determine the mode of the first model, the mode of the first model is used to determine the first model, and the first model is formed based on the structural data and the parameter data;
  • the second communication interface used for outputting the channel state information reference signal CSI-RS
  • the second communication interface is further configured to acquire second channel state information CSI
  • the second logic circuit is further configured to determine the first CSI according to the first model and the second CSI, where the data amount of the second CSI is smaller than the data amount of the first CSI.
  • the second communication device can determine the mode of the first model according to the actual communication situation, and the methods for determining the structural data and parameter data of the first model are different between different modes, which improves the flexibility of the solution and improves the efficiency of the solution.
  • the methods for determining the structural data and parameter data of the first model are different between different modes, which improves the flexibility of the solution and improves the efficiency of the solution.
  • the second communication device acquires the second CSI, which can reduce the overhead of CSI feedback compared to the first CSI. Further, the second communication device may determine the first CSI based on the second CSI fed back by the first communication device and the first model, obtain complete channel information, and improve communication performance.
  • the embodiments of the present invention further provide a first implementation manner of the fourth aspect:
  • the second logic circuit is further configured to determine the mode of the first model according to the first parameter, where the first parameter includes one or more of channel transmission scenarios, channel transmission uplink and downlink resources, and device hardware parameters.
  • the second communication device can predefine the mode of the channel learning model according to the first parameter, and the first parameter includes the channel transmission scenario, the channel transmission uplink and downlink resources, and the device capability of the first communication device. It interacts with the signaling of the second communication device, and while reducing the signaling interaction, the mode of the channel learning model is determined according to the actual communication situation, and the channel state information is fed back.
  • the second communication device may determine the mode of the channel learning model according to the first parameter. That is, different modes of the channel learning model can be determined according to different values or different situations of the first parameter.
  • the determination of the mode of the channel learning model takes into account the conditions of different first parameters, which can be more adaptable to different transmission scenarios, uplink and downlink resources, and device capabilities, meet the needs of different situations, improve the accuracy of determining the first model, and improve communication. performance.
  • the embodiment of the present invention also provides a second implementation manner of the fourth aspect:
  • the second communication interface is further configured to acquire mode request information, where the mode request information is used to request to determine the mode of the first model.
  • the first communication device sends mode request information to the second communication device, so that the second communication device determines the channel learning model mode, and the mode request information may include one or more modes of the channel learning model.
  • the information is used for reference by the channel state feedback device, so that the second communication device can determine the channel learning model mode according to its actual communication situation and also consider the actual communication situation where the first communication device is located, which improves the flexibility of the scheme. At the same time, the rationality is further improved.
  • the embodiment of the present invention also provides a third implementation manner of the fourth aspect:
  • the second communication interface is further configured to output mode indication information, where the mode indication information is used to indicate the mode of the first model.
  • the mode indication information of the second communication device can be used to determine the channel learning model mode, and the device capability requirements of the first communication device are relatively low, and the first communication device only needs to receive the indication information of the second communication device. Determining the mode of the channel learning model improves the flexibility of the scheme.
  • the embodiment of the present invention further provides a fourth implementation of the fourth aspect:
  • the second communication interface is also used for outputting structural data and parameter data of the first model.
  • the second communication device can output the structural data and parameter data of the channel learning model to the first communication device. , so that the first communication device builds a channel learning model according to the acquired structural data and parameter data, so as to determine the channel state information after reducing the amount of data.
  • the second communication device may determine structural class data and parameter class data of the channel learning model.
  • the second communication device may determine the channel learning model based on the previous channel information, for example, the first communication device in a cell or a certain area may use the same channel learning model, which reduces the complexity of determining the channel learning model.
  • the embodiment of the present invention further provides a fifth implementation of the fourth aspect:
  • the second communication interface is further used to obtain structural data and parameter data of the first model.
  • the second communication device can obtain the structural data and parameter data of the channel learning model from the first communication device. data, so that the second communication device builds a channel learning model according to the acquired structural data and parameter data, so as to determine the channel state information after restoring the data amount.
  • the complexity of training the second communication device to determine the channel learning model is reduced, and the energy saving of the second communication device is realized.
  • the first communication device determines the structural data and the parameter data
  • the embodiment of the present invention further provides a sixth implementation of the fourth aspect:
  • the second communication interface is also used to output the structural data of the first model
  • the second communication interface is further configured to acquire parameter class data, where the parameter class data is determined according to the structural class data.
  • the mode for determining the channel learning model is that the second communication device determines the structure class data and the first communication device determines the parameter class data
  • the first communication device learns the model structure class according to the channel determined by the second communication device.
  • the data further determines the parameter class data of the channel learning model.
  • the second communication device determines the structure class data, which reduces the requirement on the capability of the first communication device and increases the implementability of the solution.
  • the second communication device can determine the structural data according to the previous channel information, the second communication device can grasp the complexity of the channel learning model, reduce the complexity of the first communication device training to determine the parameter data, and avoid the first communication device.
  • Communication devices need to train different channel learning models for performance under structured data multiple times.
  • the first communication device trains the matched parameter class data based on the structural class data determined by the second communication device, which can reduce the complexity of the first communication device training and determine the channel learning model and realize the energy saving of the first communication device.
  • the parameter type data determined by the first communication device may be that the first communication device trains the channel learning model according to the channel environment, so that the parameter type data more matches the current channel environment, improves the accuracy of the channel learning model, and improves the communication performance.
  • the embodiment of the present invention further provides a seventh implementation of the fourth aspect:
  • the second communication interface is also used to obtain structural data of the first model
  • the second logic circuit is further configured to determine the parameter class data of the first model according to the structure class data;
  • the second communication interface is also used for outputting parameter data.
  • the second communication device learns the model structure class according to the channel determined by the first communication device.
  • the data further determines the parameter data of the channel learning model, which reduces the amount of data that needs to be processed by the second communication device to construct the channel learning model in the actual solution implementation process, which is more suitable for the situation where the first communication device has strong capabilities.
  • the first communication device can determine the structural data of the channel learning model based on the capabilities of the first communication device, the first communication device can grasp the complexity of the channel learning model, and reduce the complexity of the second communication device in determining the parameter data , to achieve energy saving of the second communication device.
  • the parameter-type data determined by the second communication device may be parameter-type data determined by the second communication device according to the previous channel information of the channel learning model, for example, the first communication device in a cell or a certain area may use the same
  • the channel learning model reduces the complexity of determining the channel learning model.
  • the embodiment of the present invention further provides an eighth implementation manner of the fourth aspect:
  • a second communication interface used to obtain the structural class data
  • a second logic circuit configured to determine first feedback configuration information, where the first feedback configuration information has a corresponding relationship with structural data
  • the second communication interface is further configured to acquire parameter class data, where the parameter class data is determined according to the first feedback configuration information and the structure class data.
  • the first communication device feeds back the structural data and the parameter data to the second communication device.
  • Hierarchical feedback can be performed, and the second communication device correspondingly determines the first feedback configuration information according to the structure data, so that the first communication device determines the parameter data according to the first feedback configuration information and the structure data.
  • the output structural data and parameter data cause greater pressure on the channel, more reasonable parameter data is further determined according to the first feedback configuration information.
  • the embodiment of the present invention further provides the ninth implementation manner of the fourth aspect:
  • the second communication interface is specifically used to obtain the first sub-data of the parameter-type data, where the first sub-data is included in the parameter-type data, and the first sub-data is determined according to the first feedback configuration information and the structure-type data.
  • the second communication device can construct a channel learning model in the case of receiving part of the parameter data, which reduces the impact on the channel when the first communication device feeds back the parameter data.
  • pressure which reduces the occupation of channel resources when feeding back CSI.
  • the embodiment of the present invention further provides the tenth implementation of the fourth aspect:
  • the second communication interface is specifically used to obtain the second sub-data of the parameter-type data, and the second sub-data is included in the parameter-type data;
  • a second logic circuit specifically configured to determine second feedback configuration information, where the second feedback configuration information has a corresponding relationship with the second sub-data of the parameter-type data
  • the second communication interface is further configured to acquire third sub-data of the parameter data, the third sub-data is determined according to the second feedback configuration information and the second sub-data, and the third sub-data is included in the Describe the parameter class data.
  • the first communication device can output the parameter data to the second communication device in a hierarchical manner.
  • the second sub-data of the data and the second feedback configuration information generated after the second communication device acquires the second sub-data determines the third sub-data of the parameter-type data, which can reduce the data volume of the third sub-data and improve the third sub-data at the same time.
  • the accuracy of the data improves the flexibility and accuracy of the solution implementation.
  • the embodiment of the present invention further provides the eleventh implementation of the fourth aspect:
  • the second communication interface is specifically used to obtain parameter data according to the time characteristic of the parameter data.
  • the parameter data can contain multiple specific parameter data, and the optimal feedback cycle corresponding to each parameter data is different, the parameters can be obtained according to the optimal feedback cycle of each parameter data, reducing the need for Channel resources occupied for CSI feedback.
  • the embodiment of the present invention further provides a twelfth implementation manner of the fourth aspect:
  • the second communication interface is specifically used to output structural data
  • a second logic circuit specifically configured to determine first feedback configuration information, where the first feedback configuration information has a corresponding relationship with structural data
  • the second logic circuit is further configured to determine parameter type data according to the first feedback configuration information and structure type data;
  • the second communication interface is also used for outputting parameter data.
  • the first communication device can obtain the structural data and parameter data output by the second communication device in stages, wherein the parameter data is determined by the structural data and the first feedback configuration relationship, which avoids outputting the structure at one time.
  • the class data and the parameter class data cause greater pressure on the channel, more reasonable parameter class data is further determined according to the first feedback configuration information of the second communication device.
  • the embodiment of the present invention further provides the thirteenth implementation manner of the fourth aspect:
  • the second communication interface is specifically used to output the first sub-data of the parameter-type data, the first sub-data is included in the parameter-type data, and the first sub-data is determined according to the first feedback configuration information and the structure-type data.
  • the parameter data of the output part of the second communication device can enable the first communication device to construct a channel learning model, which reduces the need for the second communication device to feedback parameter data.
  • the pressure of the channel reduces the occupation of channel resources when feeding back CSI.
  • the embodiment of the present invention further provides the fourteenth implementation manner of the fourth aspect:
  • the second logic circuit is specifically configured to train the first model according to the first feedback configuration information and the structural data, and determine the parameter data of the first model.
  • the second communication device can retrain the channel learning model according to the first feedback configuration information, thereby determining the parameter data that is more suitable for the current channel environment, which improves the second communication device. Accuracy of parametric data as determined by communications equipment.
  • the embodiment of the present invention further provides the fifteenth aspect of the fourth aspect Implementation:
  • a second communication interface specifically configured to output second sub-data of the parameter-type data, the second sub-data being included in the parameter-type data
  • a second logic circuit specifically configured to determine second feedback configuration information, where the second feedback configuration information has a corresponding relationship with the second sub-data of the parameter-type data
  • the second logic circuit is further configured to determine the third sub-data of the parameter-type data according to the second feedback configuration information and the second sub-data;
  • the second communication interface is also used for outputting the third sub-data.
  • the second communication device can output the parameter class data to the first communication device in a hierarchical manner.
  • the second sub-data of the data and the second feedback configuration information generated after the second communication device acquires the second sub-data determines the third sub-data of the parameter-type data, which can reduce the data volume of the third sub-data and improve the third sub-data.
  • the accuracy of the data improves the flexibility and accuracy of the solution implementation.
  • the embodiment of the present invention further provides the sixteenth aspect of the fourth aspect Implementation:
  • the second communication interface is used for outputting the parameter data according to the time characteristic of the parameter data.
  • the parameter output can be carried out according to the optimal feedback cycle of each parameter data, reducing the number of parameters. Channel resources occupied when feeding back CSI.
  • any of the eighth, ninth, tenth, eleventh, twelfth, thirteenth, fourteenth, fifteenth, and sixteenth types based on the fourth aspect also provides a seventeenth embodiment of the fourth aspect:
  • the second logic circuit is specifically configured to determine at least one of the resources, the number of bits, and the feedback information corresponding to the parameter type data of the first model according to the first feedback configuration information and the structure type data.
  • the first communication device can determine the configuration information corresponding to the parameter data according to the first feedback configuration information determined by the second communication device, and the configuration information can be determined according to different channel states.
  • One or more of these options improve the flexibility and applicability of this solution.
  • a fifth aspect of the embodiments of the present invention provides a first channel state information measurement feedback system, including:
  • the first processing unit is used to determine the mode of the first model, the mode of the first model is used to determine the first model, the first model is formed based on the structural data and the parameter data, the first model is a channel learning model, the channel The learning model can process the channel state information according to the configuration parameters, thereby reducing the amount of channel state information data;
  • a first obtaining unit configured to obtain a channel state information reference signal CSI-RS, where the CSI-RS is used to determine the first channel state information CSI;
  • the first processing unit is further configured to determine a second CSI according to the first model and the first CSI, where the data volume of the second CSI is smaller than the data volume of the first CSI;
  • the first output unit is used for outputting the second CSI.
  • the embodiment of the present invention also provides a first implementation manner based on the fifth aspect:
  • the second processing unit is configured to determine a mode of the first model according to a first parameter, where the first parameter includes one or more of channel transmission scenarios, channel transmission uplink and downlink resources, and device capabilities.
  • the embodiment of the present invention also provides a second implementation manner of the fifth aspect:
  • the second output unit is configured to output mode request information, where the mode request information is used to request to determine the mode of the first model.
  • the embodiment of the present invention further provides a third implementation manner of the fifth aspect:
  • the second acquiring unit is configured to acquire mode indication information, where the mode indication information is used to indicate the mode of the first model.
  • the embodiment of the present invention further provides a fourth implementation of the fifth aspect:
  • the third acquiring unit is configured to acquire structural data and parameter data of the first model.
  • the embodiment of the present invention further provides a fifth implementation of the fifth aspect:
  • the third output unit is used for outputting structural data and parameter data of the first model.
  • the embodiment of the present invention further provides a sixth implementation of the first aspect:
  • the third obtaining unit is also used to obtain the structural data of the first model
  • a third processing unit configured to determine the parameter class data of the first model according to the structural class data
  • the third output unit is also used for outputting parameter data.
  • the embodiment of the present invention further provides a seventh implementation of the first aspect:
  • the third output unit is also used to output the structural data of the first model
  • the third obtaining unit is further configured to obtain parameter class data, where the parameter class data is determined according to the structure class data.
  • the embodiment of the present invention further provides an eighth implementation manner of the fifth aspect:
  • the fourth output unit used for outputting structural data
  • a fourth obtaining unit configured to obtain first feedback configuration information, where the first feedback configuration information has a corresponding relationship with structural data
  • a fourth processing unit configured to determine parameter type data according to the first feedback configuration information and structure type data
  • the fourth output unit is further configured to output parameter data.
  • the embodiment of the present invention further provides the ninth implementation manner of the fifth aspect:
  • the fourth output sub-unit is used for outputting the first sub-data of the parameter-type data, the first sub-data is included in the parameter-type data, and the first sub-data is determined according to the first feedback configuration information and the structure-type data.
  • the embodiment of the present invention further provides the tenth implementation manner of the fifth aspect:
  • the fifth processing unit is configured to train the first model according to the first feedback configuration information and the structural data, and determine the parameter data of the first model.
  • the embodiment of the present invention further provides the eleventh implementation of the fifth aspect:
  • a fifth output unit used for outputting the second sub-data of the parameter-type data, the second sub-data being included in the parameter-type data;
  • a fifth obtaining unit configured to obtain second feedback configuration information, where the second feedback configuration information has a corresponding relationship with the second sub-data of the parameter data
  • a fifth processing unit further configured to determine, according to the second feedback configuration information and the second sub-data, third sub-data of the parameter-type data, where the third sub-data is included in the parameter-type data;
  • the fifth output unit is also used for outputting the third sub-data.
  • the embodiment of the present invention also provides the twelfth implementation of the fifth aspect:
  • the fifth output subunit is used for outputting the parameter data according to the time characteristic of the parameter data.
  • the embodiment of the present invention further provides a thirteenth implementation manner of the fifth aspect:
  • the third acquisition subunit is used to acquire structural data
  • the third obtaining subunit is further configured to obtain first feedback configuration information, where the first feedback configuration information has a corresponding relationship with structural data;
  • the third acquiring subunit is further configured to acquire parameter type data, where the parameter type data is determined according to the first feedback configuration information and the structural type data.
  • the embodiment of the present invention further provides the fourteenth implementation manner of the fifth aspect:
  • the fifth acquiring subunit is used for acquiring the first subdata of the parameter type data, the first subdata is included in the parameter type data, and the first subdata is determined according to the first feedback configuration information and the structure type data.
  • the embodiment of the present invention further provides a fifteenth implementation of the fifth aspect :
  • the fifth acquisition subunit is also used to acquire the second subdata of parameter class data, and the second subdata is included in the parameter class data;
  • the fifth obtaining subunit is further configured to obtain second feedback configuration information, where the second feedback configuration information has a corresponding relationship with the second sub-data of the parameter data;
  • the fifth acquisition subunit is further configured to acquire third subdata of parameter data, where the third subdata is determined according to the second feedback configuration information and the second subdata, and the third subdata is included in the parameter data.
  • the embodiment of the present invention further provides the sixteenth implementation manner of the fifth aspect :
  • the fifth acquiring subunit is further configured to acquire the parameter data according to the time characteristic of the parameter data.
  • the embodiment of the present invention also provides a seventeenth implementation of the fifth aspect:
  • the fifth processing subunit is configured to determine at least one of the second type of data resource, the number of bits, and the feedback information of the first model according to the first feedback configuration information and the structure type data.
  • a sixth aspect of the embodiments of the present invention provides a second channel state information measurement feedback system, including:
  • the sixth processing unit is used to determine the mode of the first model, and the mode of the first model is used to determine the first model, and the first model is formed based on structural data and parameter data;
  • a sixth output unit configured to output the channel state information reference signal CSI-RS
  • a sixth obtaining unit configured to obtain the second channel state information CSI
  • the sixth processing unit is further configured to determine the first CSI according to the first model and the second CSI, where the data amount of the second CSI is smaller than the data amount of the first CSI.
  • the embodiment of the present invention also provides a first implementation manner of the sixth aspect:
  • a seventh processing unit configured to determine a mode of the first model according to a first parameter, where the first parameter includes one or more of channel transmission scenarios, channel transmission uplink and downlink resources, and device hardware parameters.
  • the embodiment of the present invention also provides a second implementation manner of the sixth aspect:
  • a seventh acquiring unit configured to acquire mode request information, where the mode request information is used to request to determine the mode of the first model.
  • the embodiment of the present invention further provides a third implementation manner of the sixth aspect:
  • the seventh output unit is configured to output mode indication information, where the mode indication information is used to indicate the mode of the first model.
  • the embodiment of the present invention also provides a fourth implementation of the sixth aspect:
  • the eighth output unit is used for outputting structural data and parameter data of the first model.
  • the embodiment of the present invention also provides a fifth implementation of the sixth aspect:
  • the eighth acquiring unit is configured to acquire structural data and parameter data of the first model.
  • the embodiment of the present invention also provides a sixth implementation of the sixth aspect:
  • an eighth output unit further configured to output structural data of the first model
  • the eighth obtaining unit is further configured to obtain parameter class data, where the parameter class data is determined according to the structural class data.
  • the embodiment of the present invention also provides a seventh implementation of the sixth aspect:
  • the eighth obtaining unit is also used to obtain the structural data of the first model
  • an eighth processing unit configured to determine parameter class data of the first model according to the structural class data
  • the eighth output unit is also used for outputting parameter data.
  • the embodiment of the present invention further provides an eighth implementation manner of the sixth aspect:
  • the ninth acquisition unit is used to acquire structural data
  • a ninth processing unit configured to determine first feedback configuration information, where the first feedback configuration information has a corresponding relationship with structural data
  • the ninth acquiring unit is further configured to acquire parameter type data, where the parameter type data is determined according to the first feedback configuration information and the structure type data.
  • the embodiment of the present invention further provides the ninth implementation manner of the sixth aspect:
  • the ninth acquiring sub-unit is specifically configured to acquire the first sub-data of the parameter-type data, the first sub-data is included in the parameter-type data, and the first sub-data is determined according to the first feedback configuration information and the structure-type data.
  • the embodiment of the present invention also provides the tenth implementation of the sixth aspect:
  • the tenth acquisition unit is specifically used to acquire the second sub-data of the parameter-type data, and the second sub-data is included in the parameter-type data;
  • a tenth processing unit specifically configured to determine second feedback configuration information, where the second feedback configuration information has a corresponding relationship with the second sub-data of the parameter type data;
  • the tenth acquiring unit is further configured to acquire third sub-data of the parameter data, the third sub-data is determined according to the second feedback configuration information and the second sub-data, and the third sub-data is included in the parameter-type data .
  • the embodiment of the present invention also provides the eleventh implementation of the sixth aspect:
  • the ninth acquiring subunit is specifically configured to acquire the parameter data according to the time characteristic of the parameter data.
  • the embodiment of the present invention further provides the twelfth implementation manner of the sixth aspect:
  • an eighth output subunit used for outputting structural data
  • the eighth processing subunit for determining the first feedback configuration information, and the first feedback configuration information has a corresponding relationship with the structural data
  • the eighth processing subunit is further configured to determine parameter type data according to the first feedback configuration information and structure type data;
  • the eighth output subunit is also used for outputting parameter data.
  • the embodiment of the present invention further provides the thirteenth implementation manner of the sixth aspect:
  • the eighth output sub-unit is further configured to output the first sub-data of the parameter-type data, the first sub-data is included in the parameter-type data, and the first sub-data is determined according to the first feedback configuration information and the structure-type data.
  • the embodiment of the present invention further provides the fourteenth implementation manner of the sixth aspect:
  • the eighth processing subunit is further configured to train the first model according to the first feedback configuration information and the structural data, and determine the parameter data of the first model.
  • the embodiment of the present invention further provides the fifteenth aspect of the sixth aspect Implementation:
  • an eighth output subunit used for outputting the second subdata of the parameter class data, the second subdata being included in the parameter class data
  • an eighth processing subunit further configured to determine second feedback configuration information, where the second feedback configuration information has a corresponding relationship with the second sub-data of the parameter-type data;
  • the eighth processing subunit is further configured to determine the third sub-data of the parameter-type data according to the second feedback configuration information and the second sub-data;
  • the eighth output subunit is also used for outputting the third subdata.
  • the embodiment of the present invention further provides the sixteenth aspect of the sixth aspect Implementation:
  • the eighth output subunit is used for outputting the parameter data according to the time characteristic of the parameter data.
  • any of the eighth, ninth, tenth, eleventh, twelfth, thirteenth, fourteenth, fifteenth, and sixteenth types based on the sixth aspect also provides a seventeenth embodiment of the sixth aspect:
  • the tenth processing subunit is configured to determine at least one of resources, number of bits, and feedback information corresponding to the parameter type data of the first model according to the first feedback configuration information and the structure type data.
  • the technical effect of the seventeenth embodiment of the sixth aspect is similar to the technical effect of the seventeenth embodiment of the fourth aspect, and is not described in detail here for brevity.
  • a seventh aspect of the embodiments of the present invention provides a readable storage medium on which a computer program is stored, and when the program is executed by a processor, implements the method according to any one of the first aspect and the second aspect.
  • the mode of the first model is determined, the mode of the first model is used to determine the first model, and the first model is formed based on structural data and parameter data, Obtain the channel state information reference signal CSI-RS, the CSI-RS is used to determine the first channel state information CSI, and determine the second CSI according to the first model and the first CSI, and the data amount of the second CSI is smaller than the The data amount of the first CSI is output, and the second CSI is output.
  • determining the mode of the first model according to the actual situation of the channel it is possible to reasonably arrange the channel state information feedback resources and reduce the channel state information feedback overhead.
  • 1 is a schematic diagram of channel transmission between communication devices in an embodiment of the present invention
  • 2-a is a schematic diagram of a neural network in an embodiment of the present invention.
  • 2-b is another schematic diagram of a neural network in an embodiment of the present invention.
  • 2-c is another schematic diagram of a neural network in an embodiment of the present invention.
  • FIG. 3 is a schematic diagram of a CSI acquisition process in an embodiment of the present invention.
  • 4-a is a schematic diagram of a first model mode determination process in an embodiment of the present invention.
  • 4-b is another schematic diagram of a first model mode determination process in an embodiment of the present invention.
  • FIG. 5 is another schematic diagram of a first model determination process in an embodiment of the present invention.
  • FIG. 6 is another schematic diagram of a first model determination process in an embodiment of the present invention.
  • FIG. 7 is another schematic diagram of a first model determination process in an embodiment of the present invention.
  • FIG. 8 is another schematic diagram of a first model determination process in an embodiment of the present invention.
  • 9-a is another schematic diagram of a first model determination process in an embodiment of the present invention.
  • 9-b is a schematic diagram of a hierarchical feedback in an embodiment of the present invention.
  • 9-c is another schematic diagram of a hierarchical feedback in an embodiment of the present invention.
  • FIG. 10 is another schematic diagram of a first model determination process in an embodiment of the present invention.
  • FIG. 11 is another schematic diagram of a first model determination process in an embodiment of the present invention.
  • FIG. 12 is another schematic diagram of a first model determination process in an embodiment of the present invention.
  • FIG. 13 is a schematic diagram of a process for determining a CSI measurement feedback mode in an embodiment of the present invention.
  • FIG. 14 is another schematic diagram of a process for determining a CSI measurement feedback manner in an embodiment of the present invention.
  • FIG. 15 is a schematic diagram of a first communication device in an embodiment of the present invention.
  • 16 is a schematic diagram of a second communication device in an embodiment of the present invention.
  • FIG. 17 is a schematic diagram of a first channel state information measurement feedback system in an embodiment of the present invention.
  • FIG. 18 is another schematic diagram of a first channel state information measurement feedback system in an embodiment of the present invention.
  • FIG. 19 is another schematic diagram of a first channel state information measurement feedback system in an embodiment of the present invention.
  • FIG. 20 is another schematic diagram of a first channel state information measurement feedback system in an embodiment of the present invention.
  • 21 is another schematic diagram of a first channel state information measurement feedback system in an embodiment of the present invention.
  • 22 is a schematic diagram of a second channel state information measurement feedback system in an embodiment of the present invention.
  • FIG. 23 is another schematic diagram of a second channel state information measurement feedback system in an embodiment of the present invention.
  • 24 is another schematic diagram of a second channel state information measurement feedback system in an embodiment of the present invention.
  • FIG. 25 is another schematic diagram of a second channel state information measurement feedback system in an embodiment of the present invention.
  • FIG. 26 is another schematic diagram of a second channel state information measurement feedback system according to an embodiment of the present invention.
  • LTE Long Term Evolution
  • FDD frequency division duplex
  • TDD time division duplex
  • UTMS universal mobile telecommunication system
  • WiMAX worldwide interoperability for microwave access
  • 5G 5th Generation
  • NR new radio access Technology
  • 6G 6th Generation
  • the 5G mobile communication system may include a non-standalone (NSA, non-standalone) and/or an independent network (SA, standalone).
  • SA independent network
  • HAPS high altitude platform
  • NTN non-terrestrial network
  • the technical solutions provided in this application can also be applied to machine type communication (MTC, machine type communication), long term evolution technology (LTE-M, Long Term Evolution-machine), device to device (D2D, device to device) network , Machine to Machine (M2M, machine to machine) network, Internet of Things (IoT, internet of things) network or other network.
  • MTC machine type communication
  • LTE-M Long term evolution technology
  • D2D device to device
  • M2M Machine to Machine
  • IoT Internet of Things
  • the IoT network may include, for example, the Internet of Vehicles.
  • the communication methods in the Internet of Vehicles system are collectively referred to as vehicle-to-other devices (V2X, vehicle to X, X can represent anything).
  • the V2X can include: vehicle-to-vehicle (V2V, vehicle to vehicle) communication, vehicle and vehicle Infrastructure (V2I, vehicle to infrastructure) communication, vehicle-pedestrian communication (V2P, vehicle to pedestrian) or vehicle-to-network (V2N, vehicle to network) communication, etc.
  • V2V vehicle-to-vehicle
  • V2I vehicle to infrastructure
  • V2P vehicle-pedestrian communication
  • V2N vehicle-to-network
  • the network device may be any device with a wireless transceiver function.
  • the equipment includes but is not limited to: base station, evolved Node B (eNB, evolved Node B), radio network controller (RNC, radio network controller), Node B (NB, Node B), base station controller (BSC, base station controller), base transceiver station (BTS, base transceiver station), home base station (HNB, home Node B), baseband unit (BBU, baseband unit), access point (AP) in wireless fidelity (WiFi, wireless fidelity) systems , access point), wireless relay node, wireless backhaul node, transmission point (TP, transmission point) or transmission and reception point (TRP, transmission and reception point), etc., and can also be 5G, such as NR, gNB in the system , or, a transmission point (TRP or TP), one or a group of antenna panels of a base station in a 5G system, or, a network node that constitutes a gNB or a transmission
  • 5G such
  • the gNB may include a centralized unit (CU, centralized unit) and a DU, and the gNB may also include an active antenna unit (AAU, active antenna unit).
  • the CU implements some functions of the gNB
  • the DU implements some functions of the gNB.
  • the CU is responsible for processing non-real-time protocols and services, implementing radio resource control (RRC, radio resource control), and packet data convergence layer protocol (PDCP, packet data convergence). protocol) layer functions.
  • RRC radio resource control
  • PDCP packet data convergence layer protocol
  • protocol packet data convergence layer protocol
  • the DU is responsible for processing physical layer protocols and real-time services, and implementing the functions of the radio link control (RLC, radio link control) layer, medium access control (MAC, medium access control) layer, and physical (PHY, physical) layer.
  • RLC radio link control
  • MAC medium access control
  • PHY physical
  • AAU implements some physical layer processing functions, radio frequency processing and related functions of active antennas. Since the information of the RRC layer will eventually become the information of the PHY layer, or be transformed from the information of the PHY layer, therefore, in this architecture, the higher-layer signaling, such as the RRC layer signaling, can also be considered to be sent by the DU. , or, sent by DU and AAU.
  • the network device may be a device including one or more of a CU node, a DU node, and an AAU node.
  • the CU can be divided into network devices in an access network (RAN, radio access network), and the CU can also be divided into network devices in a core network (CN, core network), which is not limited in this application.
  • RAN access network
  • CN core network
  • the network equipment provides services for the cell, and the terminal equipment communicates with the cell through transmission resources (for example, frequency domain resources, or spectrum resources) allocated by the network equipment, and the cell may belong to a macro base station (for example, a macro eNB or a macro gNB, etc.) , can also belong to the base station corresponding to the small cell, where the small cell can include: urban cell (metro cell), micro cell (micro cell), pico cell (pico cell), femto cell (femto cell), etc. , these small cells have the characteristics of small coverage and low transmission power, and are suitable for providing high-speed data transmission services.
  • a macro base station for example, a macro eNB or a macro gNB, etc.
  • the small cell can include: urban cell (metro cell), micro cell (micro cell), pico cell (pico cell), femto cell (femto cell), etc.
  • these small cells have the characteristics of small coverage and low transmission power, and are suitable for providing high-speed data transmission
  • a terminal device may also be referred to as user equipment (UE, user equipment), an access terminal, a subscriber unit, a subscriber station, a mobile station, a mobile station, a remote station, a remote terminal, a mobile device, a user terminal, Terminal, wireless communication device, user agent or user equipment.
  • UE user equipment
  • an access terminal a subscriber unit, a subscriber station, a mobile station, a mobile station, a remote station, a remote terminal, a mobile device, a user terminal, Terminal, wireless communication device, user agent or user equipment.
  • a terminal device may be a device that provides voice/data connectivity to a user, and may be referred to as a terminal for short.
  • terminals can be: mobile phone (mobile phone), drone, tablet computer (pad), computer with wireless transceiver function (such as notebook computer, PDA, etc.), mobile Internet device (MID, mobile internet device) ), virtual reality (VR, virtual reality) equipment, augmented reality (AR, augmented reality) equipment, wireless terminals in industrial control (industrial control), wireless terminals in unmanned driving (self driving), remote medical (remote medical) ), wireless terminals in smart grid, wireless terminals in transportation safety, wireless terminals in smart city, wireless terminals in smart home, Cellular phones, cordless phones, session initiation protocol (SIP, session initiation protocol) phones, wireless local loop (WLL, wireless local loop) stations, personal digital assistants (PDA, personal digital assistant), handheld devices with wireless communication capabilities, Computing equipment or other processing equipment connected to a wireless
  • wearable devices can also be called wearable smart devices, which is a general term for the intelligent design of daily wear and the development of wearable devices using wearable technology, such as glasses, gloves, watches, clothing and shoes.
  • a wearable device is a portable device that is worn directly on the body or integrated into the user's clothing or accessories.
  • Wearable device is not only a hardware device, but also realizes powerful functions through software support, data interaction, and cloud interaction.
  • wearable smart devices include full-featured, large-scale, complete or partial functions without relying on smart phones, such as smart watches or smart glasses, and only focus on a certain type of application function, which needs to cooperate with other devices such as smart phones. Use, such as all kinds of smart bracelets, smart jewelry, etc. for physical sign monitoring.
  • the terminal device may also be a terminal device in an Internet of Things (IoT, internet of things) system.
  • IoT Internet of Things
  • IoT is an important part of the development of information technology in the future. Its main technical feature is to connect items to the network through communication technology, so as to realize the intelligent network of human-machine interconnection and interconnection of things.
  • IoT technology can achieve massive connections, deep coverage, and terminal power saving through, for example, narrowband (NB, narrow band) technology.
  • NB narrowband
  • terminal equipment can also include sensors such as smart printers, train detectors, and gas stations.
  • the main functions include collecting data, receiving control information and downlink data from network equipment, and sending electromagnetic waves to transmit uplink data to network equipment.
  • FIG. 1 shows a schematic diagram of a communication system 100 suitable for the method provided by this embodiment of the present application.
  • the communication system 100 may include at least one network device, such as the network device 101 in the 5G system as shown in FIG. 1 ; the communication system 100 may also include at least one terminal device, as shown in FIG. 1
  • the terminal devices 102 to 107 may be mobile or stationary.
  • Each of the network device 101 and one or more of the end devices 102 to 107 may communicate over a wireless link.
  • Each network device can provide communication coverage for a specific geographic area and can communicate with terminal devices located within that coverage area. For example, the network device may send configuration information to the terminal device, and the terminal device may send uplink data to the network device based on the configuration information; for another example, the network device may send downlink data to the terminal device. Therefore, the network device 101 and the terminal devices 102 to 107 in FIG. 1 constitute a communication system.
  • D2D or V2X technology can be used to realize direct communication between terminal devices.
  • D2D or V2X technology can be used for direct communication between terminal devices 105 and 106 and between terminal devices 105 and 107 .
  • Terminal device 106 and terminal device 107 may communicate with terminal device 105 individually or simultaneously.
  • the terminal devices 105 to 107 can also communicate with the network device 101, respectively. For example, it can communicate directly with the network device 101, as shown in the figure, the terminal devices 105 and 106 can directly communicate with the network device 101; it can also communicate with the network device 101 indirectly, as in the figure, the terminal device 107 communicates with the network device via the terminal device 106. 101 Communications.
  • FIG. 1 exemplarily shows a network device, a plurality of terminal devices, and communication links between the communication devices.
  • the communication system 100 may include multiple network devices, and the coverage of each network device may include other numbers of terminal devices, such as more or less terminal devices.
  • the communication system 100 may include multiple network devices, and the multiple network devices may perform coordinated multi-point transmission. Exemplarily, multiple network devices may cooperate or cooperate to communicate with one terminal device. This application does not limit this.
  • Each of the above communication devices may be configured with multiple antennas.
  • the plurality of antennas may include at least one transmit antenna for transmitting signals and at least one receive antenna for receiving signals.
  • the transmitting antenna and the receiving antenna may be the same or different. Exemplarily, in the same case, one antenna can be used for both transmission and reception; in different cases, the transmission antenna and the reception antenna are different antennas.
  • each communication device additionally includes a transmitter chain and a receiver chain, which can be understood by those of ordinary skill in the art, all of which may include multiple components (eg, processors, modulators, multiplexers) related to signal transmission and reception. , demodulator, demultiplexer or antenna, etc.). Therefore, the network device and the terminal device can communicate through the multi-antenna technology.
  • the wireless communication system 100 may further include other network entities such as a network controller, a mobility management entity, and the like, which are not limited in this embodiment of the present application.
  • network entities such as a network controller, a mobility management entity, and the like, which are not limited in this embodiment of the present application.
  • the channel is the channel through which the signal is transmitted in the communication system, the transmission medium through which the signal is transmitted from the transmitting end to the receiving end, and may also include related equipment for signal transmission.
  • the base station and the terminal can send signals to each other, the signal sent by the base station to the terminal is called the downlink signal, and the signal sent by the terminal to the base station is called the uplink signal.
  • the base station is the wireless base station in the network and also the network element of the wireless access network.
  • the terminal can implement direct air interface interaction with the base station, that is, can implement signal transmission and/or reception.
  • the main driving force for the improvement of the spectral efficiency of the number of antennas is that under time division duplex (TDD, Time-division Duplex), the number of antennas is doubled, and the spectral efficiency can be increased by 50%; Under FDD (Frequency-division Duplex), the number of antennas is doubled, and the spectral efficiency can be improved by 25%.
  • TDD Time division duplex
  • FDD Frequency-division Duplex
  • the base station due to channel reciprocity under TDD, the base station can accurately obtain the channel, but there is no channel reciprocity under FDD, the base station needs to perform CSI feedback through the terminal to obtain channel information, and the CSI feedback under FDD has feedback delay, feedback Quantization, feedback overhead, etc.
  • the embodiment of the present invention proposes a CSI measurement feedback method based on a channel learning model.
  • the channel learning model It may be a model or algorithm for channel acquisition, a model or algorithm for determining channel information, a model or algorithm related to a channel, or a model or algorithm applied in a communication system, and the like.
  • the channel learning model is a neural network model
  • the neural network is mainly composed of an input layer, a hidden layer and an output layer
  • the structure of the channel learning model taking the neural network model as an example can be the dimension of the input layer, the output layer, and the One or more of layer dimension, number of hidden layers, and number of hidden neurons.
  • a symmetric neural network as shown in Figure 2-a can be used.
  • the neural network includes encoding (from M dimension to D dimension), f: R M -> R D ; decoding (from D dimension to M dimension), that is, f -1 : R D -> R M ; where M can be If it is greater than D, the training can be performed with the minimum average approximation error as the loss function.
  • the training algorithms such as backpropagation algorithm, gradient descent algorithm, etc., which are not limited in this application.
  • the average training error can be expressed as:
  • cn is an M-dimensional channel vector
  • An exemplary channel matrix may be a complex number of A*B*S dimensions.
  • A is the number of antenna ports of the network device
  • B is the number of antenna ports of the base station
  • S is the number of subcarriers.
  • M can be A*B*S, that is, the real part and the imaginary part are input independently, or, A*B*S*2, that is, the real part and the imaginary part are jointly input .
  • A is 64
  • B is 1, and S is 1, then M is 64 or 128.
  • D can be a positive integer, exemplarily 2, 4, 5, 6, 8, 16, 32 and so on.
  • the neural network shown in Figure 2-a is a 4-layer neural network structure, and the number of neurons in each layer can be gradually reduced. For example, if M is 64 and D is 4, 64-dimensional high-dimensional channel information can be reduced to 4-dimensional channel information.
  • the encoding and decoding equations may be symmetric or asymmetric, that is, the two may adopt the same structure or different structures.
  • the value (activation value) of each hidden layer neuron/output layer neuron is calculated by the neurons in the previous layer, after operations (such as weighted summation, weighted summation plus bias, etc.) and nonlinear transformation obtained.
  • the nonlinear transformation function also called activation function
  • the activation functions are Ranh function and Relu function as an example.
  • Sigmoid function It is a sigmoid function common in biology, also known as the sigmoid growth curve. In information science, the sigmoid function is often used as the threshold function of neural networks due to its mono-increasing and inverse-function mono-increasing properties. Mapping variables between 0 and 1, the sigmoid function can be represented by the following formula:
  • Tanh function It is one of the hyperbolic functions. In mathematics, the Tanh function is derived from the basic hyperbolic function, hyperbolic sine function and hyperbolic cosine function. The Tanh function can be represented by the following formula:
  • Relu function used to calculate the output of hidden layer neurons.
  • the Relu function can be represented by the following formula:
  • a variant of the Relu activation function which can be a leaky linear rectification function (leakly Relu, LRelu), a leaky random linear rectification function (random leaky Relu, RRelu), or a parameter linear rectification function (parameter Relu, PRelu), etc.
  • the LRelu function can be expressed by the following formula:
  • the PRelu function can be represented by the following formula:
  • ⁇ is a variable that can be learned by backpropagation.
  • the RRelu function can be expressed by the following formula:
  • each transformation can include weighted summation plus bias and nonlinear transformation operations.
  • H g(X*W+b).
  • W is a weight matrix or a weight vector, referred to as a weight
  • b is a bias vector or a bias matrix, referred to as a bias
  • g() is an activation function.
  • X is a matrix of 1*x and the number of neurons in the hidden layer is h
  • W can be a matrix of dimension x*h
  • b can be a matrix of dimension 1*h.
  • h 1 g 1 (x 1 *w 1,1,1 +x 2 *w 1,1,2 +b 1,1 ),
  • h 2 g 1 (x 1 *w 1,2,1 +x 2 *w 1,2,2 +b 1,2 ),
  • h 50 g 1 (x 1 *w 1,50,1 +x 2 *w 1,50,2 +b 1,50 ).
  • the function g 1 may be an activation function, such as a Sigmoid function, a Tanh function or a Relu function.
  • y 1 g 2 (h 1 *w 2,1,1 +h 2 *w 2,1,2 +...+h 50 *w 2,1,50 +b 2,1 ),
  • y 2 g 2 (h 1 *w 2,2,1 +h 2 *w 2,2,2 +...+h 50 *w 2,2,50 +b 2,2 ),
  • y 3 g 2 (h 1 *w 2,3,1 +h 2 *w 2,3,2 +...+h 50 *w 2,3,50 +b 2,3 ),
  • y 4 g 2 (h 1 *w 2,4,1 +h 2 *w 2,4,2 +...+h 50 *w 2,4,50 +b 2,4 ).
  • the function g 2 may be an activation function, such as a Sigmoid function, a Tanh function or a Relu function.
  • the configuration parameters of the channel learning model involved in this step can be transformation algorithm, weight vector, weight matrix, bias vector, bias matrix, activation function, input layer dimension, output layer dimension, hidden layer number, hidden layer One or more of the number of neurons in the layer, etc.
  • the first communication device shown in the following embodiments may be replaced by a component (such as a chip or a system of chips, etc.) configured in the first communication device.
  • the second communication device shown in the following embodiments may also be replaced with a component (eg, a chip or a chip system, etc.) configured in the second communication device.
  • the embodiments shown below do not specifically limit the specific structure of the execution body of the method provided by the embodiment of the present application, as long as the program that records the code of the method provided by the embodiment of the present application can be executed to execute the method provided by the embodiment of the present application.
  • the method only needs to communicate.
  • the execution subject of the method provided by the embodiment of the present application may be the first communication device or the second communication device, or, the first communication device or the second communication device can call the program and execute the program. functional module.
  • the first communication device mentioned in the following embodiments may be a terminal device, or may be a component (such as a chip or a chip system, etc.) configured in the terminal device.
  • the second communication device may be a network device, or may be a component (such as a chip or a chip system, etc.) configured in the network device.
  • the first communication device may be a network device, or may be a chip or a chip system or other components configured in the network device.
  • the second communication device may be a terminal device, or may be a component such as a chip or a chip system configured in the terminal device.
  • the first communication device may be a terminal device, or may be a chip or a chip system or other components configured in the terminal device.
  • the second communication device may be a terminal device, or may be a component such as a chip or a chip system configured in the terminal device.
  • the first communication device may be a network device, or may be a chip or a chip system or other components configured in the network device.
  • the second communication device may be a network device, or may be a chip or a chip system or other components configured in the network device.
  • the channel learning model mentioned in the following embodiments may include a first channel learning model and a second channel learning model.
  • the first channel learning model may refer to a dimensionality reduction or encoding model
  • the second channel learning model may refer to a restoration or decoding model.
  • the first channel learning model is deployed on the side of the first communication device, and the second channel learning model is deployed on the side of the second communication device.
  • the first channel learning model and the second channel learning model are deployed on the side of the first communication device, and the first channel learning model and the second channel learning model are deployed on the side of the second communication device.
  • the channel learning model mentioned in the following embodiments may refer to the first channel learning model and/or the second channel learning model if it is not clearly specified whether it is the first channel learning model or the second channel learning model.
  • the channel learning model training mentioned in the embodiments of the present application may also be referred to as channel learning training for short.
  • the channel learning model training may include at least one of the following: determining a first channel learning model, determining a second channel learning model, determining first channel information, determining second channel information, and the like.
  • one or more rows and one column or more columns in the table may be used in practical applications, exemplarily at least one row and at least one column.
  • FIG. 3 takes the first communication device as the terminal and the second communication device as the base station as an example for illustration.
  • the embodiment of the present invention provides a Schematic flowchart of CSI acquisition.
  • the following embodiment provides a method for acquiring CSI, and the method may be used as an independent embodiment or combined with other embodiments. Specifically, the embodiment of the present application does not limit this.
  • the method shown in FIG. 3 may include steps 301 to 307 , and each step in the method will be described in detail below.
  • the base station determines the mode and the first model of the first model
  • the terminal determines the mode and the first model of the first model
  • the terminal and the base station may determine the mode and the first model of the first model.
  • 301 may be performed before 302, or 302 may be performed after 301, or the determination process of the terminal and the base station may be performed alternately, which is not limited in this application.
  • the first model may be a channel learning model, and determining the first model includes determining structural data and parameter data of the first model.
  • the structural data and parameter data are described below:
  • the structural data of the channel learning model may include one or more of the input dimension, the feature encoding dimension, the number of bits quantized and encoded, the number of neural network layers, and the number of neurons.
  • the parameter class data of the channel learning model may include one or more of weight matrices, bias vectors, quantization parameters, quantization methods, activation functions, and the like.
  • the structure of the channel learning model may include a 2-layer data network, the dimension of the input layer (also referred to as the number of neurons in the input layer) is 2, and the hidden layer The dimension (also called the number of neurons in the hidden layer) is 50, the dimension of the output layer (also called the number of neurons in the output layer) is 4, and the parameter data of the channel learning model can include the weight matrix W1 and W2, bias vectors b1 and b2, etc.
  • an exemplary real number range is (-4, 4): if equal-spaced quantization is employed.
  • the quantization interval can be 1, and the quantization value can be -4,-3,-2,-1,0,1,2,3.
  • the quantization interval can be 0.5, and the quantization value can be -3.5,-3,-2.5,-2,-2.5,-2,-1.5,-1,-0.5,0,0.5,1 , 1.5, 2, 2.5, 3, 3.5.
  • the quantization rate can be parameter type data.
  • the structural data in the channel learning model may also include one or more of the structure of the model, the dimension of the model, the operations in the model, the functions in the model, and the like.
  • the parameter class data in the channel learning model may include one or more of variables in the model, parameters in the model, and the like.
  • the structure of the model includes one or more of the structures in which the model is a neural network, principal component analysis, automatic coding, and the like.
  • the structure of the model also includes the characteristics of the input information of the model, the characteristics of the output information of the model, and the like.
  • the feature of the input information of the model may refer to the operation of the high-dimensional channel information as the input information.
  • the operation may refer to one or more of de-averaging, normalization operation, discrete Fourier transform, delay-angle domain transform, or separation of real and imaginary parts, and the like.
  • the input information can be a combination of real and imaginary parts, independent real parts, independent imaginary parts, or delay-angle domain information, etc. This operation can also be referred to as data preprocessing.
  • the features of the output information of the model may refer to the information features after encoding or dimension reduction.
  • the information after encoding or dimensionality reduction may refer to the information after normalization operation, the information after discrete Fourier transform, the information after time-delay-angle domain transformation, the information of the combination of real and imaginary parts, Or information after separation of real and imaginary parts.
  • the structure of the model is a convolutional neural network
  • the real part and the imaginary part respectively correspond to a convolutional neural network
  • the input information corresponding to the channel learning model 1 is the real part
  • the encoded output information is the real part
  • the input information corresponding to the channel learning model 2 is the imaginary part
  • the encoded output information is the imaginary part.
  • the configuration parameters of the two neural networks may be the same or different.
  • the channel learning model may include one or more models. Illustratively, there is a channel learning model 1 for the real part and a channel learning model 2 for the imaginary part.
  • Determining the channel learning model may be determining structural class data and/or parameter class data of one or more channel learning models.
  • the dimension of the model can be one or more of the number of layers in the model, the dimension of each layer, the dimension of input/hidden/output, etc. For example, it is determined that the dimension of the model is an N-layer convolutional neural network, the dimension of the input layer is N1, and the dimension of the output layer is N2. Among them, N, N1, N2 are positive integers.
  • the operations in the model can be one or more of linear operations, nonlinear operations, etc.
  • the operation to determine the model is a linear operation, and the operation can also be measured or characterized by complexity.
  • the functions in the model may be mathematical operations, logical operations, etc., illustratively one or more of addition, subtraction, multiplication and division, weighted summation, weighted summation plus bias, activation functions, and the like.
  • the functions that determine the model are the weighted sum plus bias and the Relu activation function.
  • the function can be determined separately for each layer in the model, or the same function can be used for one or more layers.
  • the variables in the model may refer to information about parameters involved in the model, exemplarily the number of parameters, the value range of the parameters, the values of the parameters, the type of parameters, and so on. Parameters can be constants or variables. For example, it is determined that the weight matrix of the model is W, the bias matrix is b, the variable value of the activation function ⁇ , and so on.
  • determining the channel learning model by the base station may mean that the base station determines the overall channel learning model for encoding and decoding, or may refer to the base station determining the channel learning model corresponding to decoding.
  • the terminal determining the channel learning model may refer to the terminal determining the overall channel learning model of encoding and decoding, or may refer to the terminal determining the channel learning model corresponding to the encoding.
  • the structural data of the channel learning model in this embodiment of the present application may refer to the structural data of the overall channel learning model for encoding and decoding, or may refer to the structural data of the encoded channel learning model, or may be Refers to the structural data of the decoded channel learning model, which is not specifically limited in this application.
  • the parameter class data of the channel learning model in the embodiment of the present application may refer to the parameter class data of the overall channel learning model for encoding and decoding, or may refer to the parameter class data of the encoded channel learning model, or may be Refers to the parameter class data of the decoded channel learning model, which is not specifically limited in this application.
  • the channel learning model in this application may include feature encoding and feature decoding, and may also include bit quantization encoding and bit quantization decoding.
  • the coded channel learning model in this application may be a channel learning model including feature coding and bit quantization coding.
  • the decoded channel learning model in this application may be a channel learning model including feature decoding and bit quantization decoding.
  • the structural data can be the number of layers of the neural network
  • the parameter data can be the weight matrix of the neural network
  • weight matrix for each layer, that is, there may be two weight matrices in total.
  • weight matrix for each layer, that is, there may be two weight matrices in total.
  • the number of weight matrices of the neural network can be determined according to the number of layers of the neural network, that is, the parameter data of the channel learning model can be determined according to the structural data of the channel learning model.
  • An exemplary structure may be the dimension of the neural network, and the configuration parameters may be the weight matrix and bias matrix of the neural network.
  • the weight matrix may be an N1*N2-dimensional matrix, and the matrix may include N1*N2 variable values.
  • the bias matrix is N2-dimensional, and the matrix can include N2 variable values.
  • the dimension of the weight matrix and/or the bias matrix of the neural network can be determined according to the dimension of the neural network, that is, the parameter data of the channel learning model can be determined according to the structural data of the channel learning model.
  • Exemplary structures may be of the type of neural networks, and may exemplarily be partially connected neural networks, or fully connected neural networks.
  • different neural network structures can correspond to different configuration parameters (such as parameter data) specifications.
  • the partially connected bias matrix in Figure 2-c can be four 1*3 matrices, that is, a total of 12 parameters.
  • the fully connected bias matrix in Figure 2-c is a 4*7 matrix, that is, a total of 28 parameters.
  • determining the mode of the first model may refer to determining the mode of structural data and parameter data of the first model, may also refer to a method of determining the first model, or may refer to a method of determining the first model.
  • the mode of the first model may be at least one of the following four types: the first communication device determines the structural data and parameter data of the first model, and the second communication device determines the structural data and the parameter data of the first model.
  • the first communication device determines the structure class data of the first model, and the second communication device determines the parameter class data of the first model, the second communication device determines the parameter class data of the first model, and the first communication device determines the first model.
  • Structural class data for the model is
  • the mode of the first model can be at least one of the following four types: the base station determines the structural data and parameter data of the first model, and the terminal determines the structure of the first model Class data and parameter class data, the base station determines the structure class data of the first model and the terminal determines the parameter class data of the first model, the base station determines the parameter class data of the first model and the terminal determines the structure class data of the first model.
  • the mode determination process of the first model may include a determination method in which a signaling interaction process exists and a determination method in which there is no signaling interaction process.
  • the following embodiments provide a method for the first communication device and/or the second communication device to determine the mode of the first model.
  • the manner of determining the mode of the first model may be used as an independent embodiment, or may be combined with other embodiments, which is not limited in this application.
  • the first communication device and/or the second communication device may determine the mode of the first model in at least one of the following manners.
  • One or more of the following methods for determining the first mode may be used alone or in combination, which is not specifically limited in this application.
  • the following description takes the first communication device as a terminal and the second communication device as a base station as an example, and other situations are similar, and details are not repeated here.
  • the first communication device and the second communication device may determine the mode of the first model according to the first parameter, where the first parameter may include one or more of channel transmission scenarios, channel transmission uplink and downlink resources, device capabilities, and location information .
  • the corresponding relationship may be predefined by a protocol, or may be notified by the second communication device to the first communication device through signaling, which is not specifically limited in this application.
  • the first communication device and/or the second communication device may store the correspondence. Furthermore, the first communication device and/or the second communication device may determine the mode of the first model according to the first parameter and the corresponding relationship between the first parameter and the mode of the first model.
  • the first communication device and/or the second communication device determines the mode of the first model according to the scene.
  • the base station and/or the terminal may determine the mode of the channel learning model according to the scene where it is located, where the scene may refer to an indoor scene, an outdoor scene, a suburban scene, an urban scene, a macro site scene, a micro site scene, a daytime scene, and a nighttime scene.
  • the scene may refer to an indoor scene, an outdoor scene, a suburban scene, an urban scene, a macro site scene, a micro site scene, a daytime scene, and a nighttime scene.
  • the scene may refer to an indoor scene, an outdoor scene, a suburban scene, an urban scene, a macro site scene, a micro site scene, a daytime scene, and a nighttime scene.
  • the scene may refer to an indoor scene, an outdoor scene, a suburban scene, an urban scene, a macro site scene, a micro site scene, a daytime scene, and a nighttime scene.
  • the base station can determine the structure and configuration parameters of the channel learning model according to big data analysis, that is, the base station can determine that the indoor scenario can be a mode in which the base station determines the structure and configuration parameters.
  • the base station can determine that the outdoor scene can be a mode in which the base station determines the structure and the terminal determines the configuration parameters.
  • different scenarios may correspond to different modes of the first model.
  • the first communication device and/or the second communication device may determine the mode of the first model according to the scene and the corresponding relationship between the scene and the mode of the first model.
  • the mode of the first model can be determined without signaling interaction, and the CSI measurement feedback can be realized.
  • the modes and scenarios of the first model can be represented by codes, the corresponding relationship can be as shown in Table 1.
  • Table 1 shows an example of the first mapping relationship.
  • the first mapping relationship may be one or more lines in the following table, and the first mapping relationship may be predefined by a protocol, or may be notified by the second communication device to the first communication device through signaling, for example, a base station The terminal is notified through signaling, which is not specifically limited in this embodiment of the present application.
  • Mode 1 to Mode M in the present invention may refer to the modes of the first model. Exemplarily, it may include at least one of the following modes: Mode 1 is a mode in which the base station determines structural data and parameter data, Mode 2 is a mode in which the base station determines structural data, and a terminal determines a mode in which parameter data is determined. Mode 3 is a mode in which the terminal determines Modes of structural data and parameter data. Mode 4 is a mode in which the terminal determines the structural data and the base station determines the parameter data.
  • Scenario 1 to Scenario N may refer to indoor scenarios, outdoor scenarios, suburban scenarios, urban scenarios, macro site scenarios, micro cell scenarios, daytime scenarios, night scenarios, sunny day scenarios, rainy day scenarios, direct path scenarios, indirect path scenarios, and quasi-scenarios. At least one of the direct path scene, etc.
  • mode 1 can determine structural data and parameter data for the base station
  • scenario 2 corresponds to an outdoor scenario with a complex channel environment
  • mode 2 can determine structural data for the base station
  • the terminal determines the parameter class data
  • the base station may determine that the mode of the first model is mode 1 according to the first mapping relationship; when the base station determines that the current scene is scene 2, the base station may determine the first model according to the first mapping relationship.
  • the mode of a model is mode 2; and so on.
  • the terminal may determine that the mode of the first model is mode 1 according to the first mapping relationship; when the terminal determines that the current scene is scene 2, the terminal may determine that the current scene is scene 2 according to the first mapping relationship. Determine the mode of the first model as mode 2; and so on.
  • the first communication device and/or the second communication device determine the mode of the first model according to uplink and downlink resources.
  • the base station and/or the terminal may determine the mode of the first model according to the uplink and downlink resources of the channel in which it is located.
  • the mode of the first model needs to be consistent with the actual uplink and downlink resources. Line resources match.
  • the base station in the mode in which the base station determines structural data and parameter data, the base station will notify the terminal of the structural data and parameter data of the channel learning model.
  • downlink signaling notification is required, that is, it can be applied to downlink resources. Scenarios with sufficient or large downlink resources.
  • the terminal In the mode in which the terminal determines the structural data and parameter data, the terminal will inform the base station of the structural data and parameter data of the channel learning model.
  • uplink signaling notification is required, which can be applied to the situation that the uplink resources are sufficient or the uplink resources are sufficient. larger scene.
  • different uplink and downlink resources may correspond to different modes of the first model.
  • the first communication device and/or the second communication device may determine the mode of the first model according to the uplink and downlink resources and the correspondence between the uplink and downlink resources and the mode of the first model.
  • the situation of the uplink and downlink resources and the mode of the first model are established in a corresponding relationship, it can be as shown in Table 2.
  • Table 2 shows an example of the second mapping relationship.
  • the second mapping relationship may be one or more lines in the following table, and the second mapping relationship may be predefined by a protocol, or may be notified by the second communication device to the first communication device through signaling, for example, a base station The terminal is notified through signaling, which is not specifically limited in this embodiment of the present application.
  • Mode 1 to Mode M in the present invention may refer to the modes of the first model. Exemplarily, it may include at least one of the following modes: Mode 1 is a mode in which the base station determines structural data and parameter data, Mode 2 is a mode in which the base station determines structural data, and a terminal determines a mode in which parameter data is determined. Mode 3 is a mode in which the terminal determines Modes of structural data and parameter data. Mode 4 is a mode in which the terminal determines the structural data and the base station determines the parameter data.
  • N1-Nn are real numbers
  • M1-Mn are real numbers
  • N1:M1-Nn:Mn represents a proportional relationship between downlink resources and uplink resources.
  • the uplink resources and/or downlink resources may also be side link resources, or may also be other communication resources, which are not specifically limited in this application.
  • the base station needs to send the structural data and parameter data of the first model to the terminal, which is suitable for the case where the downlink resources are sufficient or the downlink resources are relatively occupied. big situation.
  • the base station will notify the terminal of the structural data and parameter data of the first model.
  • downlink signaling notification is required, that is, it can be applied to downlink resources. Scenarios with sufficient or large downlink resources.
  • the terminal will notify the base station of the structural data and parameter data of the first model.
  • uplink signaling notification is required, that is, it can be applied to the situation that the uplink resources are sufficient or the uplink resources are sufficient.
  • the first communication device and/or the second communication device determines the mode of the first model according to device capabilities.
  • the base station and/or the terminal may determine the mode of the first model according to the device capability.
  • the device capability may refer to whether the first communication device has structural data for determining the first model. and/or the capability of parameter data, it may also refer to whether the first communication device has the ability to train the first model, or it may refer to whether the first communication device has the capability related to channel learning, specifically, this application does not Do limit.
  • whether the terminal has the ability to determine the structure and/or configuration parameters of the channel learning model, whether the terminal has the ability to train the channel learning model, and whether the terminal has the ability related to channel learning.
  • the first communication device may report the device capability to the second communication device, that is, the first communication device may send the device capability to the second communication device through signaling.
  • the signaling in this embodiment of the present application may refer to high-level signaling, for example, radio resource control (RRC, radio resource control) signaling, medium access control (MAC, medium access control) signaling, etc.
  • RRC radio resource control
  • MAC medium access control
  • DCI downlink control information
  • UCI uplink control information
  • SCI sidelink control information
  • the first communication device and/or the second communication device may determine the mode of the first model according to the device capability and the corresponding relationship between the device capability and the mode of the first model.
  • a corresponding relationship is established between the device capability and the mode of the first model, as shown in Table 3.
  • An example of the third mapping relationship is shown in Table 3.
  • the third mapping relationship can be one or more lines in the following table.
  • the third mapping relationship can be predefined by the protocol, or the second communication device can pass If the signaling informs the first communication device, the base station exemplarily informs the terminal through signaling, which is not specifically limited in this embodiment of the present application.
  • C is a positive integer, that is, the device capability is the C-th capability, and the device capability C is a device capability.
  • the mode of the first model is Mode 1, and the base station determines the structural data and parameter data.
  • the terminal can inform the base station of the structural data and parameter data of the first model, and may select mode 3 to determine the first model mode, that is, the terminal Structural class data and parameter class data of the first model are determined.
  • the first communication device and/or the second communication device determines the mode of the first model according to the location of the device.
  • the device may refer to a first communication device and/or a second communication device.
  • the geographic location where the device is located may be three-dimensional coordinates, two-dimensional coordinates, positioning data, and the like.
  • the following takes the device as the terminal as an example. If the geographical location where the terminal is located is area 1, the base station and/or the terminal may determine the mode of the first model corresponding to area 1; if the geographical location where the terminal is located is area 2, the base station and/or the terminal may determine the mode corresponding to area 2 The mode of the corresponding first model.
  • the geographic location information in this application may also be related information after encryption or virtualization or other operations are performed on the geographic location information based on the rules of terminal privacy protection.
  • the information of the geographic location may also be virtual information, or code information or the like. Specifically, this application does not limit this.
  • the first communication device and/or the second communication device may determine the mode of the first model according to the position of the device and the corresponding relationship between the position of the device and the mode of the first model.
  • Table 4 shows an example of the fourth mapping relationship.
  • the fourth mapping relationship may be one or more lines in the following table, and the fourth mapping relationship may be predefined by a protocol, or may be notified by the second communication device to the first communication device through signaling, for example, a base station The terminal is notified through signaling, which is not specifically limited in this embodiment of the present application.
  • Device location model Device Location 1 or Zone 1 Mode 1 Device Location 2 or Zone 2 Mode 2 Device Location 3 or Zone 3 Mode 3 ... ... device location p or region p Mode M
  • p is a positive integer, that is, the p-th position or position area is the area p.
  • the first communication device and/or the second communication device determine the mode of the first model according to the scenario, at least two items of uplink and downlink resources, device capability, and device location.
  • Table 5 shows that in the communication process of the embodiment of the present invention, the way of determining the first model mode by the base station and/or the terminal may be based on the consideration of scenarios, uplink and downlink resource conditions, device capabilities, and device locations. determined after a number of factors.
  • the main factors for determining the first model mode may be uplink and downlink resource conditions and scenarios, and when A certain factor has a special situation.
  • the terminal does not have the determination of structural data and parameter data, it can be considered to perform CSI measurement feedback through the mode in which the base station determines the structural data and parameter data.
  • the base station and/or terminal may determine that the mode of the first model is mode 1.
  • the base station and/or terminal may determine that the mode of the first model is mode 2.
  • the base station and/or terminal may determine that the mode of the first model is mode 3.
  • the first communication device and/or the second communication device may correspond to the mode of the first model according to the scenario, at least two of the uplink and downlink resources and the device capability and the scenario, at least two of the uplink and downlink resources and the device capability
  • the relationship determines the schema of the first model. Exemplarily, if a corresponding relationship is established between at least two of the above four factors and the mode of the first model, it can be shown in Table 5. Table 5 shows an example of the fifth mapping relationship.
  • the fifth mapping relationship may be one or more lines in the following table, and the fifth mapping relationship may be predefined by a protocol, or may be notified by the second communication device to the first communication device through signaling, for example, a base station The terminal is notified through signaling, which is not specifically limited in this embodiment of the present application.
  • Table 5 shows that in the communication process of the embodiment of the present invention, the way of determining the first model mode by the base station and/or the terminal may be after considering various factors in the scenario, uplink and downlink resource conditions, and device capabilities. definite.
  • the main factors for determining the first model mode may be uplink and downlink resource conditions and scenarios, and when There is a special case for a certain factor, for example, when the terminal does not have the determination of structural data and parameter data, it can be considered to determine the first model through the mode in which the base station determines the structural data and parameter data, and determine the first model according to the first model.
  • the model performs CSI measurement feedback.
  • the base station and/or terminal may determine that the mode of the first model is mode 1.
  • the base station and/or terminal may determine that the mode of the first model is mode 2.
  • the base station and/or terminal may determine that the mode of the first model is mode 3.
  • the determination of the first model mode is realized without signaling interaction, which saves channel resources, and through the analysis of scenarios, uplink and downlink resources.
  • the comprehensive consideration of the situation and equipment capability and equipment location improves the accuracy and practicability of the first model and improves the communication performance.
  • the base station sends a channel state information reference signal to the terminal;
  • step 303 may be after steps 301 and/or 302, or may be before steps 301 and/or 302. Specifically, the present application does not limit the order of steps 301, 302, and 303.
  • the base station sends the channel state information reference signal to the terminal, or the base station outputs the channel state information reference signal.
  • the channel state information reference signal is used to determine the first channel state information.
  • the base station sends the channel state information reference signal to the terminal, so that the terminal obtains downlink channel information according to the channel state information reference signal.
  • TDD due to the mutuality of channels, both communication parties can obtain channel information accurately
  • FDD taking the base station and the terminal as an example, the base station needs the terminal to perform CSI feedback to obtain channel information.
  • the channel state information reference signal in this embodiment of the present application may also be a downlink reference signal, or a downlink signal, exemplarily a synchronization signal, a broadcast signal, and a demodulation reference signal , cell-level reference signal, terminal reference signal, phase tracking reference signal, tracking reference signal, data channel, control channel and other signals or one or more of the channels, the channel state information reference signal can be used to obtain channel state information signal of.
  • CSI-RS Channel state information-reference signal
  • Channel state information reference signal may also be a downlink reference signal, or a downlink signal, exemplarily a synchronization signal, a broadcast signal, and a demodulation reference signal , cell-level reference signal, terminal reference signal, phase tracking reference signal, tracking reference signal, data channel, control channel and other signals or one or more of the channels.
  • the base station may send CSI measurement feedback resources to the terminal.
  • the CSI measurement feedback resources in this embodiment may refer to physical uplink shared channel resources (PUSCH, physical uplink shared channel), physical uplink control Channel resources (PUCCH, physical uplink control channel), or physical feedback channel resources (PFCH, physical feedback channel), sidelink feedback resources (SFCH, sidelink feedback channel), etc.
  • the CSI measurement feedback resources may refer to resources related to CSI feedback, that is, the CSI measurement feedback resources may carry or carry information related to CSI measurement feedback.
  • the terminal may feed back information related to the CSI measurement feedback on the CSI measurement feedback resource, and the base station may receive the information related to the CSI measurement feedback on the CSI measurement feedback resource.
  • the information related to the CSI measurement feedback may include one of structural data of the first model, parameter data of the first model, low-dimensional CSI, channel amplitude information, channel phase information, rank information, precoding information, channel quality information, and the like. item or multiple items.
  • the terminal acquires a channel state information reference signal, and the channel state information reference signal is used to determine the first CSI;
  • step 304 may be after steps 301 and/or 302, or may be before steps 301 and/or 302.
  • the present application does not limit the order of steps 301, 302, and 304.
  • the terminal acquires the channel state information reference signal, or the terminal receives the channel state information reference signal.
  • the terminal may receive and/or acquire the channel state information reference signal, and the terminal may acquire the first channel information according to the channel state information reference signal.
  • the first channel information may be downlink channel information; for another example, when the first communication device is a network device, the first channel information may be uplink channel information.
  • the first channel information may be uplink channel information, or the first channel information may be uplink channel information and downlink channel information, and the first communication device may be based on the uplink and downlink channel information. Part of the reciprocity, determining the first channel learning model and/or the second channel learning model according to the uplink channel information and the downlink channel information.
  • the first channel information may be downlink channel information, or the first channel information may be uplink channel information and downlink channel information, and the first communication device may be based on the uplink and downlink channel information.
  • the partial reciprocity of the first channel learning model and/or the second channel learning model is determined according to the uplink channel information and the downlink channel information.
  • the channel information in this embodiment of the present application may refer to channel state information, and similarly, the channel state information may also be referred to as channel information for short. This application does not limit this.
  • the terminal determines the second CSI based on the first model and the first CSI;
  • the data amount of the second CSI is smaller than the data amount of the first CSI.
  • the first model may include a first channel learning model and a second channel learning model.
  • the first channel learning model can be used to determine the second channel information based on the first channel information, and the data volume of the second channel information is smaller than the data volume of the first channel information. Therefore, it can also be said that the first channel learning model is used to determine the second channel information.
  • One channel information is compressed to obtain second channel information.
  • the data amount of the channel information may refer to the dimension of the channel information.
  • the number of antenna ports of the transmitting end (for example, it may be the first communication device or the second communication device) is A2, and the number of antenna ports of the receiving end (for example, it may be the first communication device or the second communication device) is A2.
  • the first channel information between the transmitting end and the receiving end may be a matrix of A2*A3 dimension, and the data amount of the first channel information may be represented by A2*A3. If the elements in the matrix of the first channel information are complex numbers, and the real part and the imaginary part of each element are expressed separately, the data amount of the first channel information can also be expressed as A2*A3*2.
  • the data amount of the second channel information may be represented by B2.
  • the data amount of the channel information may also refer to the amount of information included in the channel information, and the like.
  • the first channel information may be regarded as the input of the first channel learning model
  • the second channel information may be regarded as the output of the first channel learning model.
  • the data volume of the first channel information may be the input information dimension
  • the data volume of the second channel information may be the output information dimension.
  • the second channel information is used to obtain third channel information through the second channel learning model, and the data amount of the third channel information and the first channel information is the same or similar.
  • the third channel information may be used for data transmission.
  • the second communication device may determine scheduling information for data transmission, or determine precoding for data transmission, etc., according to the third channel information.
  • the third channel information may also be the first channel information.
  • the terminal obtains the channel state information of the downlink channel through the channel state information reference signal sent by the base station, and performs dimension reduction and/or compression on the channel state information of the downlink channel based on the first model, thereby determining the first CSI.
  • each transformation may include weighted summation plus bias and nonlinear transformation operations .
  • exemplary take the operation of transforming from X to H as an example, it can be H g(X*W+b), where W is a weight matrix or a weight vector, referred to as a weight, and the dimension of W can be X A matrix of dimensions of dimension*H.
  • b is a bias vector or a bias matrix, referred to as bias, the dimension of b can be a matrix of dimensions of 1*H, and g() is an activation function.
  • the input layer is 2-dimensional
  • the 1*2 matrix X [x_axis, y_axis]
  • the hidden layer is 50-dimensional
  • the 1*50 matrix H [h1, h2, . . . , h50].
  • weighted summation plus bias and nonlinear transformation g1 can be the following operations
  • H X*W1+b1, where W1 is a 2*50 matrix, which can be called a weight matrix, and b1 is a 1*50 matrix, which can be called a bias.
  • h1 g1(x_axis*w1,1+y_axis*w1,2+b1,1)
  • h2 g1(x_axis*w2,1+y_axis*w2,2+b1,2)
  • h50 g1(x_axis*w50,1+y_axis*w50,2+b1,50)
  • the function g1 may be an activation function, exemplarily a function of sigmoid, tanh, relu and the like.
  • weighted summation plus bias and nonlinear transformation g2 can be the following operations
  • Y H*W2+b2, where W2 is a 50*4 matrix, and b2 is a 1*4 matrix.
  • y2 g2(x_axis*w2,1+y_axis*w2,2+b2,2)
  • y50 g2(x_axis*w50,1+y_axis*w50,2+b2,50)
  • the function g2 can be an activation function.
  • the parameter class data of the first model can be a transformation algorithm, a weight vector, a weight matrix, a bias vector, a bias matrix, an activation function, an input layer dimension, an output layer dimension, a hidden layer number, and a hidden layer neuron. one or more of numbers, etc.
  • the structural data of the first model may be one or more items of activation function, dimension of input layer, dimension of output layer, number of hidden layer, number of neurons in hidden layer, etc.
  • the terminal obtains downlink channel information, that is, the first CSI according to the channel state information reference signal, and processes the first CSI based on the first model to obtain the second CSI, and the data volume of the second CSI is smaller than the first CSI
  • the channel state information is represented by less information, which improves the communication performance.
  • the terminal sends the second CSI to the base station; correspondingly, the base station receives the second CSI sent by the terminal.
  • the terminal sends the second CSI to the base station, or the terminal outputs the second CSI.
  • the base station receives the second CSI sent by the terminal, or the base station may acquire the second CSI.
  • the base station determines the first CSI according to the second CSI based on the first model.
  • the base station determines the first CSI according to the first model and the second CSI.
  • the data amount of the second CSI is smaller than that of the first CSI.
  • the first model may include a first channel learning model and a second channel learning model.
  • the second channel learning model can be used to determine the first channel information based on the second channel information, and the data volume of the second channel information is smaller than the data volume of the first channel information. Therefore, it can also be said that the second channel learning model is used to determine the first channel information.
  • the second channel information is decompressed to obtain the first channel information.
  • the data amount of the channel information may refer to the dimension of the channel information.
  • the number of antenna ports of the transmitting end (for example, it may be the first communication device or the second communication device) is A 2
  • the antenna ports of the receiving end for example, it may be the first communication device or the second communication device
  • the number is A 3
  • the first channel information between the transmitting end and the receiving end may be an A 2 *A 3 -dimensional matrix
  • the data amount of the first channel information may be represented by A 2 *A 3
  • the elements in the matrix of the first channel information are complex numbers, and the real part and the imaginary part of each element are expressed separately, the data amount of the first channel information can also be expressed as A 2 *A 3 *2.
  • the data amount of the second channel information may be represented by B 2 .
  • the data amount of the channel information may also refer to the amount of information included in the channel information, and the like.
  • the second channel information may be regarded as the input of the second channel learning model
  • the first channel information may be regarded as the output of the second channel learning model.
  • the data volume of the second channel information may be the input information dimension
  • the data volume of the first channel information may be the output information dimension.
  • the second channel information is used to obtain third channel information through the second channel learning model, and the data amount of the third channel information and the first channel information is the same or similar.
  • the third channel information may be used for data transmission.
  • the second communication device may determine scheduling information for data transmission, or determine precoding for data transmission, etc., according to the third channel information.
  • the third channel information may also be the first channel information.
  • the base station receives the second CSI sent by the terminal, and determines the first CSI according to the second CSI based on the first model.
  • the base station may determine the first CSI according to the second CSI and the first model sent by the terminal, which may be, for example, the original CSI.
  • the first model is a symmetric neural network as shown in FIG. 2-a
  • the decoding of the second CSI by the base station side corresponds to the encoding of the first CSI by the terminal side.
  • the equations of encoding and decoding may be symmetric or asymmetric, that is, the two may adopt the same structure or different structures.
  • an embodiment of the present application provides a communication method that can be performed by a first communication device. That is, the following method is described from the perspective of the first communication device, and the method may include at least one of S310' to S340':
  • the first communication device determines the mode of the first model and the first model. Specifically, for the description of this step, reference may be made to the description in 302 above, which is not described in detail here for brevity. Optionally, the present application does not limit the order of step S310' and step S320'.
  • the first communication device acquires a channel state information reference signal, and the channel state information reference signal is used to determine the first CSI.
  • the description of this step can be described in 304 above, and for brevity, it will not be described in detail here.
  • the present application does not limit the order of step S310' and step S320'.
  • the first communication device determines the second CSI according to the first model and the first CSI, and the data amount of the second CSI is smaller than the data amount of the first CSI. Specifically, the description of this step can be described in 305 above, and for the sake of brevity, it will not be described in detail here.
  • the first communication device outputs the second CSI.
  • the description in the above method 306 which is not described in detail here for brevity.
  • an embodiment of the present application provides a communication method that can be performed by a second communication device. That is, the following method is described from the perspective of the second communication device, and the method may include at least one of S310" to S330":
  • the second communication device determines the mode and the first model of the first model. Specifically, the description of this step may refer to the description in 301 above, and for brevity, it will not be described in detail here.
  • the sequence of S310" and step S320" is not limited.
  • the second communication device outputs the channel state information reference signal.
  • the description of this step may refer to the description in 303 above, and for brevity, it will not be described in detail here.
  • the order of S320" is not limited.
  • the second communication device obtains the second CSI. Specifically, the description of this step can be described in the above 306 , which is not described in detail here for brevity.
  • the second communication device determines the first CSI according to the first model and the second CSI, and the data amount of the second CSI is smaller than the data amount of the first CSI.
  • the description of this step can refer to the description in the above method 307, in order to For brevity, it will not be described in detail here.
  • both parties in the communication comprehensively determine the mode of the first model according to the actual communication scenario, uplink and downlink resource conditions, and device capabilities, and then perform data processing on CSI based on the model.
  • the influence of the channel state on data transmission the base station restores the received CSI to obtain the real CSI, and determines the mode of the first model according to the actual situation of the channel, so as to reasonably arrange the channel state information feedback resources and reduce the channel state information feedback overhead.
  • the embodiment of the present invention further introduces the case where signaling interaction is required for the determination of the first model mode.
  • signaling interaction is required for the determination of the first model mode.
  • FIG. 4-a please refer to FIG. 4-a.
  • the following embodiments provide a method for the first communication device and/or the second communication device to determine the mode of the first model.
  • the manner of determining the mode of the first model may be used as an independent embodiment, or may be combined with other embodiments, which is not limited in this application.
  • the first communication device and/or the second communication device may determine the mode of the first model in at least one of the following manners.
  • One or more of the following methods for determining the first mode may be used alone or in combination, which is not specifically limited in this application. The following only takes the first communication device as a terminal and the second communication device as a base station as an example for description, and other situations are similar, and details are not repeated here.
  • a schematic diagram of the first model mode determination process in the embodiment of the present invention includes:
  • the terminal sends mode request information to the base station; correspondingly, the base station receives the mode request information sent by the terminal.
  • the terminal sending the mode request information to the base station may also refer to the terminal outputting the mode request information.
  • the base station receiving the mode request information sent by the terminal may also refer to the base station acquiring the mode request information.
  • the mode request information is used to request to determine the mode of the first model.
  • the base station may receive mode request information sent by the terminal, where the mode request information refers to information for requesting to determine the first model mode.
  • the mode request information is request information, which is only used to request the base station to determine the mode of the first model.
  • the mode request information does not carry the mode information of the first model.
  • the mode request information may refer to a bit information or a sequence information.
  • the request information is used to request the base station to determine the mode of the first model.
  • the base station may send mode indication information to the terminal, where the mode indication information is used to indicate the mode of the first model. That is, the mode request information may refer to requesting the base station to send mode indication information.
  • the schema request information includes schema information of the first model.
  • the terminal may send information about the mode of the first model to the base station, or the terminal may send information about the mode of the candidate first model that the terminal considers suitable to the base station, or the terminal may send the base station Information on the mode of the first model that is considered the most applicable.
  • the information of the mode of the first model may be used to indicate one or more modes of the first model.
  • the base station may determine the mode indication information according to the mode information sent by the terminal. That is, the mode information sent by the terminal can be used as reference information for the base station to determine the mode.
  • the mode request information includes identification information.
  • the identification information is used to indicate the mode of the first model.
  • the identification information has a corresponding relationship with the mode.
  • the information of the mode with the first model in the mode request information may be identifier information, for example, identifier 1 corresponds to mode 1, and identifier 1 corresponds to mode 1.
  • 2 represents mode 2, or the symbol a corresponds to mode 1, the symbol b corresponds to mode 2, and so on.
  • the mode request information sent by the terminal to the base station can determine the mode of the structural data and parameter data of the first model for the base station. In the process, if the above mode corresponds to mode 1, then the mode request information only needs to carry the identification information of mode 1.
  • the base station determines the mode of the first model.
  • the base station may determine the mode of the first model according to the mode request information sent by the terminal, and may consider the suggestion of the terminal to improve the rationality of the base station in determining the mode of the first model.
  • the base station may determine the mode of the first model according to the mode request information sent by the terminal, or may not determine the mode of the first model according to the mode request information sent by the terminal, that is, the mode of the first model determined by the base station may be It is the same as or different from the mode information contained in the mode request information sent by the terminal.
  • the mode request information sent by the terminal to the base station may contain one or more mode information or mode identifiers, when the mode request information only contains one mode information or mode identifier, and the base station determines according to the mode request information.
  • the mode of the first model is the same as the mode indicated by the mode request information, the base station may not send mode indication information to the terminal, when the mode request information sent by the terminal contains a variety of mode information or the mode and mode request determined by the base station When the modes contained in the information are different, the base station needs to send mode indication information to the terminal, so that the terminal acquires accurate mode information of the first model.
  • the method of determining the mode of the first model according to the request of the terminal can comprehensively consider the suggestion of the terminal, improve the rationality of the base station to determine the first model, and determine the method of determining the first model, It can be applied to various complex environments such as various scenarios, various resource conditions, various equipment capabilities, etc., to improve the accuracy of the channel learning model and improve the communication performance.
  • the base station sends the mode indication information to the terminal for further introduction.
  • the base station sends the mode indication information to the terminal for further introduction.
  • Fig. 4-b please refer to Fig. 4-b.
  • the following embodiments provide a method for the first communication device and/or the second communication device to determine the mode of the first model.
  • the manner of determining the mode of the first model may be used as an independent embodiment, or may be combined with other embodiments, which is not limited in this application.
  • the first communication device and/or the second communication device may determine the mode of the first model in at least one of the following manners.
  • One or more of the following methods for determining the first mode may be used alone or in combination, which is not specifically limited in this application. The following only takes the first communication device as a terminal and the second communication device as a base station as an example for description, and other situations are similar, and details are not repeated here.
  • Another schematic diagram of the first model mode determination process in the embodiment of the present invention includes:
  • the base station sends mode indication information to the terminal; accordingly, the terminal receives the mode indication information sent by the base station;
  • the base station sends the mode indication information to the terminal, which may also refer to the base station outputting the mode indication information.
  • the terminal receiving the mode indication information sent by the base station may also refer to the terminal acquiring the mode indication information.
  • the mode indication information is used to indicate the mode of the first model.
  • the base station may send mode indication information to the terminal, so that the terminal determines the mode of the first model.
  • the base station sends the mode indication information to the terminal:
  • the base station receives the mode request information sent by the terminal, which contains multiple mode information, or when the mode information determined by the base station is different from the mode information included in the mode request information, the base station can send the mode determination information to the terminal. , so that the terminal determines the mode of the first model.
  • the second is that the base station receives the mode request information sent by the terminal, which contains multiple mode information, or when the mode information determined by the base station is the same as the mode information included in the mode request information, the base station can send the mode determination information to the terminal. , so that the terminal determines the mode of the first model.
  • the base station can ignore the mode information in the mode request information sent by the terminal, or when the mode request information sent by the terminal does not carry the mode information, the base station can directly send the mode indication information to the terminal.
  • the mode of the first model, the mode indication information may include the identification information of the mode of the first model, for example, identification 1 corresponds to mode 1, identification 2 represents mode 2, or identification a corresponds to mode 1, identification b corresponds to mode 2, etc.
  • the method means that the terminal directly determines the first model mode, which reduces the equipment capability requirements of the terminal.
  • the terminal determines the mode of the first model.
  • the terminal determines the mode of the first model according to the mode indication information sent by the base station, which may be to determine the mode of the first model by using the mode identification information contained in the mode indication information.
  • an embodiment of the present application provides a communication method that can be performed by a first communication device. That is, the following method is described from the perspective of the first communication device, and the method may include at least one of S410' to S440':
  • the first communication device determines the mode of the first model. Specifically, for the description of this step, reference may be made to the description in 302 above, which is not described in detail here for brevity. Optionally, this step can be omitted.
  • S420' The first communication device outputs mode request information, where the mode request information is used to request to determine the mode of the first model. Specifically, for the description of this step, reference may be made to the description in 401 above, which is not described in detail here for brevity.
  • the first communication device acquires mode indication information, where the mode indication information is used to indicate the mode of the first model.
  • the mode indication information is used to indicate the mode of the first model.
  • this step can be omitted.
  • the first communication device determines the mode of the first model. Specifically, the description of this step can be described in 404 above, and for brevity, it will not be described in detail here. Optionally, this step can be omitted.
  • an embodiment of the present application provides a communication method that can be performed by a first communication device. That is, the following method is described from the perspective of the first communication device, and the method may include at least one of S410* to S420*:
  • the first communication device acquires mode indication information, where the mode indication information is used to indicate the mode of the first model. Specifically, for the description of this step, reference may be made to the description in 403 above, which is not described in detail here for brevity.
  • the first communication device determines the mode of the first model. Specifically, the description of this step can be described in 404 above, and for brevity, it will not be described in detail here.
  • an embodiment of the present application provides a communication method that can be performed by a second communication device. That is, the following method is described from the perspective of the second communication device, and the method may include at least one of S410" to S420":
  • the second communication device determines the mode of the first model. Specifically, for the description of this step, reference may be made to the description in 301 above, which is not described in detail here for brevity.
  • the second communication device outputs the mode indication information. Specifically, for the description of this step, reference may be made to the description in 403 above, which is not described in detail here for brevity.
  • an embodiment of the present application provides a communication method that can be performed by a second communication device. That is, the following method is described from the perspective of the second communication device, and the method may include at least one of S410** to S420**:
  • the second communication device acquires mode request information. Specifically, for the description of this step, reference may be made to the description in 401 above, which is not described in detail here for brevity.
  • the second communication device outputs mode indication information.
  • this step reference may be made to the description in 403 above, which is not described in detail here for brevity.
  • this step can be omitted.
  • an embodiment of the present application provides a communication method that can be performed by a second communication device. That is, the following method is described from the perspective of the second communication device, and the method may include at least one of S410*** to S430***:
  • the second communication device acquires mode request information. Specifically, for the description of this step, reference may be made to the description in 401 above, which is not described in detail here for brevity.
  • the second communication device determines the mode of the first model. Specifically, for the description of this step, reference may be made to the description in 402 above, which is not described in detail here for brevity.
  • the second communication device outputs mode indication information. Specifically, for the description of this step, reference may be made to the description in 403 above, which is not described in detail here for brevity.
  • the base station can flexibly consider various scenarios, realize the adaptive change of the method, improve the rationality of the terminal determining the channel learning model, and determine the
  • the method of determining the channel learning model can be applied to various scenarios, various resource conditions, various equipment capabilities and other complex environments, so as to improve the accuracy of the channel learning model and improve the communication performance.
  • the communication process between the terminal and the base station is taken as an example to introduce the situation that the base station sends the mode indication information to the terminal, because in general, the equipment capability of the base station is better than that of the terminal, so the base station is used as the mode determined by the base station. one end.
  • priority rules can be established based on device capabilities. For example, if the device capabilities of terminal A are better than those of terminal B, then terminal A can be used as the end of the mode determination. However, if the device capabilities of the two devices are similar, the first model mode can be determined by performing mode negotiation between the network device that sends the CSI and the network device that receives the CSI.
  • the party with weak device capability may be set as the first communication device, and the party with strong device capability may be set as the second communication device.
  • the party sending the CSI may be the first communication device, and the party receiving the CSI may be the second communication device, and so on. Specifically, this application does not limit this.
  • the first model can be determined according to the mode, which specifically includes determining the structural data and parameter data of the first model according to the mode, mode 1-mode 4 respectively correspond to a mode of the first model, and each mode may correspond to a determination method of the first model.
  • the mode which specifically includes determining the structural data and parameter data of the first model according to the mode
  • mode 1-mode 4 respectively correspond to a mode of the first model
  • each mode may correspond to a determination method of the first model.
  • the following embodiments provide a method for the first communication device and/or the second communication device to determine the first model.
  • the manner of determining the first model may be used as an independent embodiment, or may be combined with other embodiments, which is not limited in this application.
  • the first communication device and/or the second communication device may determine the first model in at least one of the following manners.
  • One or more of the following methods for determining the first model may be used alone or in combination, which is not specifically limited in this application. In the following, only the first communication device is used as a terminal and the second communication device as a base station is used as an example for description, and other situations are similar, and details are not repeated here.
  • a schematic diagram of a first model determination process in an embodiment of the present invention includes:
  • the base station determines structural data and parameter data of the first model
  • the base station can determine structural data and parameter data according to one or more of scenarios, uplink and downlink resources, device capabilities, and device locations.
  • the structural data may be the input dimension and the output dimension
  • the parameter data may be the weight matrix and the bias matrix.
  • the base station may construct the first model according to the determined structural data and parameter data, and decompress and/or restore the CSI according to the constructed first model.
  • the base station sends parameter data and structural data to the terminal; correspondingly, the terminal receives the structural data and parameter data sent by the base station.
  • the base station sending structural data and parameter data to the terminal may also refer to the base station outputting structural data and parameter data.
  • the terminal receiving structural data and parameter data sent by the base station may also mean that the terminal obtains structural data and parameter data.
  • the terminal determines the first model according to the parameter data and the structure data
  • the terminal can receive the parameter data and structural data sent by the base station, and construct the first model according to the structural data and parameter data sent by the base station.
  • the CSI can reduce the dimension of 64-dimensional data to 4-dimensional data through the processing of the first model on the terminal side, which reduces the number of channels in the communication process. Occupation of resources.
  • an embodiment of the present application provides a communication method that can be performed by a first communication device. That is, the following method is described from the perspective of the first communication device, and the method may include at least one of S510' to S520':
  • the first communication device acquires structural data and parameter data. Specifically, for the description of this step, reference may be made to the description in 502 above, which is not described in detail here for brevity. Optionally, this step can be omitted.
  • the first communication device determines the first model according to the structural data and the parameter data. Specifically, for the description of this step, reference may be made to the description in 503 above, which is not described in detail here for brevity.
  • an embodiment of the present application provides a communication method that can be performed by a second communication device. That is, the following method is described from the perspective of the second communication device, and the method may include at least one of S510" to S520":
  • the second communication device determines the structural data and parameter data of the first model. Specifically, for the description of this step, reference may be made to the description in 501 above, which is not described in detail here for brevity.
  • the second communication device outputs structural data and parameter data. Specifically, for the description of this step, reference may be made to the description in 502 above, which is not described in detail here for brevity.
  • the base station determines the signal transmission and reception situation of structural data and parameter data.
  • the base station determines the structural data and parameter data, which can reduce the requirements for the equipment capability of the terminal, and avoid the terminal from determining the structural data and parameters.
  • the data-like process enables the terminal to save energy.
  • this embodiment is more suitable for scenarios where the channel environment is simple or stable.
  • the same first model may be used for terminals in a cell or a certain area, and the base station determines the first model based on previous channel information. Determine the first model.
  • this embodiment is more suitable for a terminal that does not have the first model training capability, and for the terminal, a mode in which the base station determines structural data and parameter data may be adopted.
  • this embodiment is more suitable for a scenario with abundant downlink resources, and can reduce the overhead of determining the uplink resources of the first model.
  • This embodiment is a process of determining the first model, and CSI measurement feedback may be performed subsequently, and the method is the same as that in Embodiments 303-307, and details are not repeated here.
  • the first model can be determined according to the mode, which specifically includes determining the structural data and parameter data of the first model according to the mode, mode 1-mode 4 respectively correspond to a mode of the first model, and each mode can correspond to a determination method of the first model.
  • the terminal determines the structural data and parameter data of the first model will be introduced. For details, please refer to Figure 6 .
  • the following embodiments provide a method for the first communication device and/or the second communication device to determine the first model.
  • the manner of determining the first model may be used as an independent embodiment, or may be combined with other embodiments, which is not limited in this application.
  • the first communication device and/or the second communication device may determine the first model in at least one of the following manners.
  • One or more of the following methods for determining the first model can be used alone or in combination, which is not specifically limited in this application. The following only takes the first communication device as a terminal and the second communication device as a base station as an example for description, and other situations are similar, and details are not repeated here.
  • a schematic diagram of a first model determination process in an embodiment of the present invention includes:
  • the terminal determines structural data and parameter data of the first model
  • the terminal can determine structural data and parameter data according to one or more of scenarios, uplink and downlink resources, device capabilities, and device location.
  • the structural data may be the input dimension and the output dimension
  • the parameter data may be the weight matrix and the bias matrix.
  • the terminal may construct a first model according to the determined structural data and parameter data, and perform CSI compression and/or dimension reduction according to the constructed first model.
  • the terminal sends structural data and parameter data to the base station; correspondingly, the base station receives the structural data and parameter data sent by the terminal.
  • the terminal sending structural data and parameter data to the base station may also refer to the terminal outputting structural data and parameter data.
  • the base station receives structural data and parameter data sent by the terminal, which may also mean that the base station obtains structural data and parameter data.
  • the base station determines the first model according to the parameter data and the structure data.
  • the terminal may send the parameter data and the structural data to the base station, so that the base station constructs the first model according to the structural data and the parameter data sent by the terminal .
  • the CSI can reduce the 64-dimensional data to 4-dimensional data through the processing of the first model on the terminal side.
  • the input dimension of the first model is 4, and the output dimension is 64, thereby realizing the process of restoring low-dimensional CSI to high-dimensional CSI and reducing the occupation of channel resources in the communication process.
  • an embodiment of the present application provides a communication method that can be performed by a first communication device. That is, the following method is described from the perspective of the first communication device, and the method may include at least one of S610' to S620':
  • the first communication device determines the structural data and parameter data of the first model. Specifically, for the description of this step, reference may be made to the description in 601 above, which is not described in detail here for brevity. Optionally, this step can be omitted.
  • the first communication device outputs structural data and parameter data. Specifically, for the description of this step, reference may be made to the description in 602 above, which is not described in detail here for brevity.
  • an embodiment of the present application provides a communication method that can be performed by a second communication device. That is, the following method is described from the perspective of the second communication device, and the method may include at least one of S610" to S620":
  • the second communication device acquires the structural data and parameter data of the first model. Specifically, for the description of this step, reference may be made to the description in 602 above, which is not described in detail here for brevity.
  • the second communication device determines the first model according to the structural data and the parameter data. Specifically, for the description of this step, reference may be made to the description in 603 above, which is not described in detail here for brevity.
  • the terminal determines the signal transmission and reception situation of the structural data and the parameter data, which can reduce the complexity of the base station training channel learning model, and the terminal determines the structural data and the parameter data.
  • the first model makes the structural data and parameter data more compatible with the current channel environment, improves the accuracy of the channel learning model, and improves the communication performance.
  • this embodiment is more suitable for a scene with a relatively complex channel environment, exemplarily an outdoor scene, because the determination of the parameter type data by the terminal side is more accurate at this time.
  • this embodiment is more suitable for a terminal with channel learning model training capability, and for the terminal, a mode in which the terminal determines structural data and parameter data may be adopted.
  • This embodiment is a process of determining the first model, and CSI measurement feedback may be performed subsequently, and the method is the same as that in Embodiments 303-307, which will not be repeated here.
  • the first model can be determined according to the mode, which specifically includes determining the structural data and parameter data of the first model according to the mode, mode 1-mode 4 respectively correspond to a mode of the first model, and each mode can correspond to a determination method of the first model.
  • the base station determines the structural data of the first model and the terminal determines the parameter data of the first model will be introduced. , see Figure 7 for details.
  • the following embodiments provide a method for the first communication device and/or the second communication device to determine the first model.
  • the manner of determining the first model may be used as an independent embodiment, or may be combined with other embodiments, which is not limited in this application.
  • the first communication device and/or the second communication device may determine the first model in at least one of the following manners.
  • One or more of the following methods for determining the first model can be used alone or in combination, which is not specifically limited in this application. The following only takes the first communication device as a terminal and the second communication device as a base station as an example for description, and other situations are similar, and details are not repeated here.
  • a schematic diagram of a first model determination process in an embodiment of the present invention includes:
  • the base station determines structural data of the first model
  • the base station can determine the structural data according to one or more of scenarios, uplink and downlink resources, device capabilities, and device locations.
  • the structural data may be input dimensions and output dimensions, so that the terminal determines parameter data according to the structural data, and constructs the first model.
  • the base station sends structural data to the terminal; correspondingly, the terminal receives the structural data sent by the base station.
  • the base station sends structural data to the terminal, which may also refer to the base station outputting structural data.
  • the terminal receiving the structural data sent by the base station may also refer to the terminal acquiring the structural data.
  • the terminal determines parameter data according to the structural data
  • the base station sends structural data to the terminal, so that the terminal determines parameter data according to the structural data.
  • the base station may determine the structure type data of the first model, and input the dimension and bit overhead exemplarily, and the parameter type data determined by the terminal may be determined by the terminal after training, exemplarily a weight matrix and a bias matrix,
  • the weight matrix realizes the dimension change of CSI
  • the bias matrix can adjust the accuracy of the matrix after the dimension change.
  • the terminal side may need to train the first model according to the channel environment, so that the parameter data can better match the current channel environment, improve the accuracy of the first model, and improve communication performance.
  • the terminal sends parameter data to the base station; correspondingly, the base station receives the parameter data sent by the terminal.
  • the terminal sending parameter data to the base station may also refer to the terminal outputting parameter data.
  • the base station receives the parameter type data sent by the terminal, which may also mean that the base station obtains the parameter type data.
  • the base station determines the first model according to the parameter class data.
  • the terminal determines parameter type data according to the structure type data sent by the base station, and sends the parameter type data to the base station, and the base station constructs the first model according to the parameter type data and the structure type data.
  • an embodiment of the present application provides a communication method that can be performed by a first communication device. That is, the following method is described from the perspective of the first communication device, and the method may include at least one of S710' to S730':
  • the first communication device acquires the structural data of the first model. Specifically, for the description of this step, reference may be made to the description in 702 above, which is not described in detail here for brevity. Optionally, this step can be omitted.
  • the first communication device determines parameter type data according to the structure type data. Specifically, for the description of this step, reference may be made to the description in 703 above, which is not described in detail here for brevity.
  • the first communication device outputs parameter data. Specifically, for the description of this step, reference may be made to the description in 704 above, which is not described in detail here for the sake of brevity.
  • an embodiment of the present application provides a communication method that can be performed by a second communication device. That is, the following method is described from the perspective of the second communication device, and the method may include at least one of S710" to S740":
  • the second communication device determines the structural class data of the first model. For a specific description of this step, reference may be made to the description in 701 above, which is not described in detail here for brevity.
  • the second communication device outputs the structural class data of the first model.
  • the structural class data of the first model For a specific description of this step, reference may be made to the description in 702 above, which is not described in detail here for brevity.
  • the second communication device acquires the parameter class data of the first model. Specifically, for the description of this step, reference may be made to the description in 704 above, which is not described in detail here for brevity.
  • the second communication device determines the first model according to the structural data and the parameter data. For a specific description of this step, reference may be made to the description in 705 above, which is not described in detail here for brevity.
  • the base station determines the structural data of the first model according to the previous channel information, grasps the complexity of the first model, reduces the complexity of terminal training, and avoids the need for the terminal to train multiple times under the structure of different channel learning models. performance.
  • the terminal trains the matched parameters based on the structure determined by the base station, which can reduce the complexity of the terminal training channel learning model.
  • the terminal can train the channel learning model according to the channel environment, so that the parameter data can better match the current channel environment, improve the accuracy of the channel learning model, and improve the communication performance.
  • this embodiment is more suitable for a scene with a relatively complex channel environment, exemplarily an outdoor scene.
  • this embodiment is more suitable for a terminal with channel learning model training capability, and for the terminal, a mode in which the terminal determines parameter-type data may be adopted.
  • this embodiment is more suitable for a scenario with abundant uplink resources, and can reduce the overhead of determining the downlink resources of the channel learning model.
  • This embodiment is a process of determining the first model, and CSI measurement feedback may be performed subsequently, and the method is the same as that in Embodiments 303-307, which will not be repeated here.
  • the first model can be determined according to the mode, which specifically includes determining the structural data and parameter data of the first model according to the mode, mode 1-mode 4 corresponds to a mode of the first model, and each mode can correspond to a determination method of the first model.
  • the base station determines the parameter data of the first model and the terminal determines the structure data of the first model. , see Figure 8 for details.
  • the following embodiments provide a method for the first communication device and/or the second communication device to determine the first model.
  • the manner of determining the first model may be used as an independent embodiment, or may be combined with other embodiments, which is not limited in this application.
  • the first communication device and/or the second communication device may determine the first model in at least one of the following manners.
  • One or more of the following methods for determining the first model can be used alone or in combination, which is not specifically limited in this application. The following only takes the first communication device as a terminal and the second communication device as a base station as an example for description, and other situations are similar, and details are not repeated here.
  • a schematic diagram of a first model determination process in an embodiment of the present invention includes:
  • the terminal determines structural data of the first model
  • the terminal can determine the structural data according to one or more of scenarios, uplink and downlink resources, device capabilities, and device locations, which reduces the time when the base station determines the parameter data. Complexity.
  • the structural data may be input dimensions and output dimensions, so that the base station determines parameter data according to the structural data and constructs the first model.
  • the terminal sends the structural data to the base station; correspondingly, the base station receives the structural data sent by the terminal.
  • the terminal sending structural data to the base station may also refer to the terminal outputting structural data.
  • the base station when the base station receives the structural data sent by the terminal, it may also mean that the base station obtains structural data.
  • the base station determines parameter data according to the structure data
  • the base station receives the structural data sent by the terminal and determines the parameter data according to the structural data.
  • the terminal may determine the structural data of the first model, exemplarily input the dimension and bit overhead, and the parameter data determined by the base station may be determined by the base station after training, exemplarily a weight matrix and a bias matrix,
  • the weight matrix realizes the dimension change of CSI
  • the bias matrix can adjust the accuracy of the matrix after the dimension change.
  • the base station can determine the parameter data according to the previous channel information and the structural data sent by the terminal, and use the previous channel information as a reference to reduce the first model deterministic complexity.
  • the base station sends parameter data to the terminal; correspondingly, the terminal receives the structure data sent by the base station.
  • the base station sends structural data to the terminal, which may also refer to the base station outputting structural data.
  • the terminal receiving the structural data sent by the base station may also refer to the terminal acquiring the structural data.
  • the terminal determines the first model according to the parameter class data.
  • the base station determines parameter type data according to the structure type data sent by the terminal, and sends the parameter type data to the terminal, so that the terminal builds a first model according to the parameter type data, and performs dimension reduction and/or compression of CSI.
  • an embodiment of the present application provides a communication method that can be performed by a first communication device. That is, the following method is described from the perspective of the first communication device, and the method may include at least one of S810' to S840':
  • the first communication device determines the structural data of the first model. Specifically, for the description of this step, reference may be made to the description in 801 above, which is not described in detail here for the sake of brevity.
  • the first communication device outputs the structural data of the first model. Specifically, for the description of this step, reference may be made to the description in 802 above, which is not described in detail here for brevity.
  • the first communication device acquires parameter data of the first model. Specifically, for the description of this step, reference may be made to the description in 804 above, which is not described in detail here for brevity.
  • the first communication device determines the first model according to the structural data and the parameter data. Specifically, for the description of this step, reference may be made to the description in 805 above, which is not described in detail here for brevity.
  • an embodiment of the present application provides a communication method that can be performed by a second communication device. That is, the following method is described from the perspective of the second communication device, and the method may include at least one of S810" to S830":
  • the second communication device acquires the structural data of the first model.
  • the description of this step may refer to the description in 802 above, which is not described in detail here for brevity.
  • this step may be omitted.
  • the second communication device determines the parameter type data according to the structure type data. Specifically, for the description of this step, reference may be made to the description in the above 803, which is not described in detail here for the sake of brevity.
  • the second communication device outputs parameter data.
  • the terminal may determine the channel learning based on the capability of the terminal
  • the structure type data of the model the terminal can grasp the complexity of the channel learning model, and reduce the complexity of the base station to determine the parameter type data.
  • the base station determines the parameter type data, which can be the base station determines the parameter type data of the channel learning model according to the previous channel information, Exemplarily, the same channel learning model may be used for a terminal in a cell or a certain area, which reduces the complexity of determining the channel learning model.
  • this embodiment is more suitable for a scene with a relatively complex channel environment, exemplarily an outdoor scene.
  • this embodiment is more suitable for a terminal with channel learning model training capability, and for the terminal, a mode in which the terminal determines structural data may be adopted.
  • this embodiment is more suitable for scenarios with abundant downlink resources, and can reduce the overhead of determining uplink resources of the channel learning model.
  • the devices at both ends of the communication determine the mode of the first model, they can determine the first model according to the mode, which specifically includes determining the structural data and parameter data of the first model according to the mode.
  • structural class data and parameter class data can be sent in stages. The following describes the situation in which the terminal sends structural class data and parameter class data in stages. Please refer to Figure 9-a for details. .
  • the following embodiments provide a method for the first communication device and/or the second communication device to determine the first model.
  • the manner of determining the first model may be used as an independent embodiment, or may be combined with other embodiments, which is not limited in this application.
  • the first communication device and/or the second communication device may determine the first model in at least one of the following manners.
  • One or more of the following methods for determining the first model can be used alone or in combination, which is not specifically limited in this application. The following only takes the first communication device as a terminal and the second communication device as a base station as an example for description, and other situations are similar, and details are not repeated here.
  • Another schematic diagram of a first model determination process in an embodiment of the present invention includes:
  • the terminal determines structural data of the first model
  • the structural data when the terminal sends the structural data and parameter data of the first model hierarchically, since the structural data of the neural network model has a certain influence on the parameter data, the structural data can be sent to the base station side to obtain the information from the base station side.
  • the structural data For the first feedback configuration information of the structure type data, more accurate feedback configuration information of the parameter type data is determined, thereby avoiding waste of feedback resources.
  • the terminal may determine the structural data of the first model according to one or more items of scenario, uplink and downlink resources, device capability, and device location.
  • the structure type data may include one or more items of Rank value, wideband amplitude, dimension, compression rate, feedback bit number, and quantization rate.
  • the terminal sends structural data to the base station; correspondingly, the base station receives the structural data sent by the terminal.
  • the terminal sending structural data to the base station may also refer to the terminal outputting structural data.
  • the base station when the base station receives the structural data sent by the terminal, it may also mean that the base station obtains structural data.
  • the base station determines the first feedback configuration information according to the structural data
  • the base station receives structural data sent by the terminal, and may determine first feedback configuration information according to the structural data, where the first feedback configuration information may include CSI measurement feedback resources, and the size of the CSI measurement feedback resources may be determined by the resource One or more of the number of blocks (RB, resource block), the number of resource elements (RE, resource element), and the number of symbols.
  • the first feedback configuration information may include CSI measurement feedback resources
  • the size of the CSI measurement feedback resources may be determined by the resource One or more of the number of blocks (RB, resource block), the number of resource elements (RE, resource element), and the number of symbols.
  • the CSI measurement feedback resources may also be referred to as CSI feedback resources for short, or may be referred to as feedback resources for short, or may be referred to as feedback time-frequency resources.
  • the first feedback configuration information is used to indicate CSI measurement feedback resources. Specifically, it may be a resource indicating feedback parameter type data.
  • the feedback configuration information may also be referred to as CSI configuration information, or simply referred to as configuration information.
  • the structure of various models can be predefined in the terminal and the base station, and the structures of different models can correspond to different usage scenarios, and can also correspond to the time-frequency resources occupied by different parameter types for data feedback.
  • the first feedback configuration The information can be used by the terminal to determine at least one of the resources, the number of bits, and the feedback information corresponding to the parameter data of the first model.
  • the base station and/or the terminal determines at least one of the resources, the number of bits, and the feedback information corresponding to the parameter type data of the first model according to the first feedback configuration information.
  • the base station determines the first feedback configuration information according to the structure type data, and may determine the first feedback configuration information in at least one of the following manners.
  • structural data and CSI measurement feedback resources have a corresponding relationship.
  • the first communication device and/or the second communication device may determine the resource corresponding to the parameter data of the first model according to the structure data and the corresponding relationship between the structure data and the CSI measurement feedback resources.
  • Table 6 shows an example of the sixth mapping relationship.
  • the sixth mapping relationship may be one or more lines in the following table, and the sixth mapping relationship may be predefined by a protocol, or may be notified by the second communication device to the first communication device through signaling, for example, a base station The terminal is notified through signaling, which is not specifically limited in this embodiment of the present application.
  • the structure type data uses a structure identifier
  • the structure identifier is used to indicate the structure type data, that is, the structure identifier is used to determine the structure of the first model.
  • the above table shows the corresponding relationship between the structure identifier and the size of the CSI measurement feedback resources.
  • the structure identifier may have a corresponding relationship with one or more of the number of symbols, the number of resource units and the number of resource blocks.
  • G is a positive integer
  • s0-ss, b0-bb, and e0-ee can be integers.
  • the structure identifier may also establish a corresponding relationship with the identifier of the CSI measurement feedback resource.
  • the structure type data has a corresponding relationship with the CSI measurement feedback resource identifier.
  • the first communication device and/or the second communication device may determine the resource corresponding to the parameter class data of the first model according to the structure class data and the correspondence between the structure class data and the CSI measurement feedback resource identifier.
  • the base station configures F CSI measurement feedback resources for the terminal through signaling.
  • F is a positive integer greater than or equal to 1.
  • the base station and/or the terminal may determine the CSI measurement feedback resource according to the structural class data of the first model.
  • Table 7 shows an example of the seventh mapping relationship.
  • the seventh mapping relationship may be one or more lines in the following table, and the seventh mapping relationship may be predefined by a protocol, or may be notified by the second communication device to the first communication device through signaling, for example, a base station The terminal is notified through signaling, which is not specifically limited in this embodiment of the present application.
  • the structure type data uses a structure identifier
  • the structure identifier is used to indicate the structure type data, that is, the structure identifier is used to determine the structure of the first model.
  • the above table shows that, since the identifier of each CSI measurement feedback resource represents the size of a CSI measurement feedback resource, in the specific implementation process, if the feedback resource identifier and CSI measurement feedback can be pre-established in both the base station and the terminal The corresponding relationship between the resource sizes, the base station can obtain the corresponding CSI measurement feedback resource identifier according to the structure identifier sent by the terminal, and send the corresponding CSI measurement feedback resource identifier to the terminal without sending the specific CSI measurement feedback resource resource.
  • the base station sends the first feedback configuration information to the terminal; correspondingly, the terminal receives the first feedback configuration information sent by the base station.
  • the base station sending the first feedback configuration information to the terminal may also mean that the base station outputs the first feedback configuration information.
  • the terminal receiving the first feedback configuration information sent by the base station may also mean that the terminal obtains the first feedback configuration information.
  • the terminal determines parameter data according to the structural data and the first feedback configuration information
  • the terminal determines the parameter type data according to the first feedback configuration information sent by the base station and the determined structure type data. If the structure type data is considered to be the first level, then the parameter type data can be considered to be the second level. Since the time-frequency resource overhead occupied by the second-level feedback is related to the first-level feedback, that is, the first-level feedback can determine the overhead of the second-level feedback. Therefore, the base station first receives the first-level feedback and then configures the time-frequency resources for the second-level feedback, and can determine the time-frequency resources for the second-level feedback according to the first-level feedback to achieve more reasonable resource allocation and avoid resource waste.
  • the determination of the parameter type data by the terminal according to the first feedback configuration information and the structure type data sent by the base station may include the following three cases:
  • the first is that the terminal determines all the parameter data according to the first feedback configuration information and the structure data.
  • the structure data determined by the current base station is inconsistent with the previously determined structure data, and the first feedback configuration information determined by the base station is used.
  • the terminal can be instructed to determine all the parameter data, and the terminal can determine the parameter data according to the first feedback configuration information and the structural data, and complete the construction of the first model according to the structural data and the parameter data, that is, determine the first model. a model.
  • the terminal determines part of the parameter type data according to the first feedback configuration information and the structure type data. For example, the structure type data determined by the current base station and a certain previously determined structure type data have the same part to a certain extent, then The first feedback configuration information determined by the base station may instruct the terminal to determine parameter data based on different parts, and the terminal can complete the construction of the first model by determining part of the parameter data.
  • the terminal may output the first sub-data of the parameter-type data, the first sub-data is included in the parameter-type data, and the first sub-data is configured according to the first feedback information and the structure class data is determined.
  • the terminal trains the first model according to the first feedback configuration information and the structural data, and determines the parameter data of the first model.
  • the parameter data and the low-dimensional CSI processed by the first model that needs to be fed back are used as the second level. Because the characteristics of CSI are related to the rank value, based on the information in the first configuration information fed back by the terminal The value of rank, the terminal can determine the feedback overhead of the low-dimensional CSI.
  • the terminal sends parameter data to the base station; correspondingly, the base station receives the parameter data sent by the terminal.
  • the terminal sending parameter data to the base station may also refer to the terminal outputting parameter data.
  • the base station receives the parameter type data sent by the terminal, which may also mean that the base station obtains the parameter type data.
  • the base station determines the first model according to the structural data and the parameter data.
  • the terminal after determining the parameter type data, the terminal sends the parameter type data to the base station, so that the base station constructs the first model according to the parameter type data.
  • an embodiment of the present application provides a communication method that can be performed by a first communication device. That is, the following method is described from the perspective of the first communication device, and the method may include at least one of S910' to S950':
  • the first communication device determines the structural data of the first model. Specifically, for the description of this step, reference may be made to the description in 901 above, which is not described in detail here for brevity.
  • the first communication device outputs the structural data of the first model. Specifically, for the description of this step, reference may be made to the description in 902 above, which is not described in detail here for brevity.
  • the first communication device acquires the first feedback configuration information. Specifically, for the description of this step, reference may be made to the description in 904 above, which is not described in detail here for brevity.
  • the first communication device determines parameter type data according to the structure type data and the first feedback configuration information. Specifically, for the description of this step, reference may be made to the description in 905 above, which is not described in detail here for brevity.
  • the first communication device outputs parameter data. Specifically, for the description of this step, reference may be made to the description in 906 above, which is not described in detail here for brevity.
  • an embodiment of the present application provides a communication method that can be performed by a second communication device. That is, the following method is described from the perspective of the second communication device, and the method may include at least one of S910" to S950":
  • the second communication device obtains the structural class data of the first model.
  • the description of this step can refer to the description in the above 802, and for brevity, it is not described in detail here.
  • this step can be omitted.
  • the second communication device determines the first feedback configuration information according to the structural data. Specifically, for the description of this step, reference may be made to the description in 903 above, which is not described in detail here for brevity.
  • the second communication device outputs the first feedback configuration information. Specifically, for the description of this step, reference may be made to the description in 904 above, which is not described in detail here for brevity.
  • the second communication device acquires parameter type data. Specifically, for the description of this step, reference may be made to the description in 906 above, which is not described in detail here for brevity.
  • the second communication device determines the first model according to the structural data and the parameter data. Specifically, for the description of this step, reference may be made to the description in 907 above, which is not described in detail here for brevity.
  • the following embodiments provide a method for the first communication device and/or the second communication device to determine the first model.
  • the manner of determining the first model may be used as an independent embodiment, or may be combined with other embodiments, which is not limited in this application.
  • the first communication device and/or the second communication device may determine the first model in at least one of the following manners.
  • One or more of the following methods for determining the first model can be used alone or in combination, which is not specifically limited in this application.
  • the following only takes the first communication device as a terminal and the second communication device as a base station as an example for description, and other situations are similar, and details are not repeated here.
  • the first communication device feeds back the CSI feedback to the second communication device in two stages.
  • the information in the first-level feedback includes structure class data of the first model, and the information in the second-level feedback includes the parameter class structure of the first model and the second CSI.
  • the feedback can be performed according to the time characteristics of the above two kinds of data.
  • the information in the first-level feedback may include one or more items of rank value, wideband amplitude, dimension, compression rate, feedback bit number, and quantization rate.
  • the information in the second level feedback may include parameter class data of the first model, and the second CSI.
  • the period of the first-level feedback may be the first period.
  • the period of the second level feedback may be the second period.
  • the value of the second period is smaller than the value of the first period.
  • the first-level feedback is only sent or received once, and the second-level feedback may be sent or received three times.
  • different levels of CSI feedback may correspond to different feedback modes, for example, the first-level feedback corresponds to periodic feedback, and the second-level feedback corresponds to aperiodic feedback; or, for example, the first-level feedback corresponds to periodic feedback.
  • the second-level feedback corresponds to semi-persistent feedback; or, exemplarily, the first-level feedback corresponds to semi-persistent feedback, the second-level feedback corresponds to aperiodic feedback, and so on.
  • the processed low-dimensional CSI (exemplarily the second CSI) can be sent together to avoid secondary occupation of channel resources.
  • the terminal divides the CSI feedback into two levels for feedback, which may be that the terminal sends structural data, and the base station sends the first feedback configuration information according to the structural data.
  • the terminal determines parameter type data according to the first feedback configuration information sent by the base station, and feeds back the parameter type data and the quantized low-dimensional CSI (exemplarily the second CSI) on the CSI measurement feedback resource indicated by the first feedback configuration information.
  • the base station can send the CSI configuration information to the terminal according to the received first-level feedback, including the time-frequency resources occupied by the second-level feedback. Because the time-frequency resource overhead occupied by the second-level feedback is related to the first-level feedback, that is, the first-level feedback can determine the overhead of the second-level feedback. Therefore, the base station first receives the first-level feedback and then configures the time-frequency resources for the second-level feedback, and can determine the time-frequency resources for the second-level feedback according to the first-level feedback to achieve more reasonable resource allocation and avoid resource waste.
  • the first communication device divides the CSI feedback into three stages and feeds it back to the second communication device.
  • the information in the first-level feedback includes the structure class data of the first model
  • the information in the second-level feedback includes the parameter class structure of the first model
  • the information in the third-level feedback includes the second CSI.
  • the feedback can be performed according to the time characteristics of the above three kinds of data.
  • the information in the first-level feedback may include one or more items of rank value, wideband amplitude, dimension, compression rate, feedback bit number, and quantization rate.
  • Information in the second level feedback may include parametric data for the first model
  • the information in the tertiary feedback may include the second CSI.
  • the period of the first-level feedback may be the third period.
  • the cycle of the second level feedback may be the fourth cycle.
  • the period of the third-level feedback may be the fifth period.
  • the value of the third period is smaller than the value of the second period, and the value of the second period is smaller than the value of the first period.
  • the first-level feedback is sent or received only once, the second-level feedback can be sent or received twice, and the third-level feedback can be sent or received four times. Second-rate.
  • different levels of CSI feedback may correspond to different feedback modes, for example, the first-level feedback corresponds to periodic feedback, the second-level feedback corresponds to aperiodic feedback, and the third-level feedback corresponds to semi-persistent feedback;
  • the first-level feedback corresponds to periodic feedback
  • the second-level feedback corresponds to semi-persistent feedback
  • the third-level feedback corresponds to aperiodic feedback
  • the first-level feedback corresponds to semi-persistent feedback
  • the second-level feedback corresponds to non-periodic feedback.
  • Periodic feedback, the third-level feedback corresponds to aperiodic feedback, etc.
  • the base station sends the second feedback configuration information to the terminal according to the received first-level feedback, including the second-level and third-level feedback time-frequency resources (including period value and/or feedback interval, etc.).
  • the time-frequency resources of the second-level feedback and the third-level feedback are determined according to the first-level feedback.
  • the time-frequency resources of the second-level feedback are determined according to the first-level feedback
  • the time-frequency resources of the third-level feedback are determined according to the second-level feedback.
  • the CSI measurement feedback resources may also be referred to as CSI feedback resources for short, or may be referred to as feedback resources for short, or may be referred to as feedback time-frequency resources.
  • the second feedback configuration information is used to indicate CSI measurement feedback resources. Specifically, it may be a resource indicating feedback of parameter type data, or may be a resource indicating feedback of the second CSI, or may be a resource indicating feedback of parameter type data and the second CSI.
  • the feedback configuration information may also be referred to as CSI configuration information, or simply referred to as configuration information.
  • the size of the CSI measurement feedback resource may include one or more of the number of symbols, the number of RBs, and the number of REs.
  • the structures of various models can be predefined in the terminal and the base station, and the structures of different models can correspond to different usage scenarios, and can also correspond to the time-frequency resources occupied by different parameter types of data feedback, or can also correspond to It corresponds to the time-frequency resources occupied by different second CSI feedback, or may also correspond to the time-frequency resources occupied by different parameter type data and second CSI feedback.
  • the second feedback configuration information may be used by the terminal to determine at least one of the resources, the number of bits, and the feedback information corresponding to the parameter data of the first model.
  • the second feedback configuration information may be used by the terminal to determine at least one of resources, number of bits, and feedback information corresponding to the second CSI.
  • the second feedback configuration information may be used by the terminal to determine at least one of the resources, the number of bits, and the feedback information corresponding to the parameter-type data of the first model, and, the resources, the number of bits, and the number of bits in the feedback information corresponding to the second CSI at least one.
  • the base station and/or the terminal determines at least one of the resources, the number of bits, and the feedback information corresponding to the parameter type data of the first model according to the second feedback configuration information.
  • the base station and/or the terminal determines at least one of the resources corresponding to the second CSI, the number of bits, and the feedback information according to the second feedback configuration information.
  • the base station determines the second feedback configuration information according to the structural data, which may be determined by at least one of the following manners.
  • structural data and CSI measurement feedback resources have a corresponding relationship.
  • the first communication device and/or the second communication device may determine the resources corresponding to the parameter data of the first model and the resources corresponding to the second CSI according to the structure data and the correspondence between the structure data and the CSI measurement feedback resources .
  • the first communication device and/or the second communication device may determine the resources corresponding to the second-level feedback and the resources corresponding to the third-level feedback according to the structure type data and the correspondence between the structure type data and the CSI measurement feedback resources.
  • Table 8 shows an example of the eighth mapping relationship.
  • the eighth mapping relationship may be one or more lines in the following table, and the eighth mapping relationship may be predefined by a protocol, or may be notified by the second communication device to the first communication device through signaling, for example, a base station The terminal is notified through signaling, which is not specifically limited in this embodiment of the present application.
  • G is a positive integer
  • s0 to ss, b0 to bb, and e0 to ee may be integers.
  • s0' ⁇ ss', b0' ⁇ bb', e0' ⁇ ee' can be integers. If a corresponding relationship is established between the structure identifier and the size of the specific feedback resource, it can be shown in Table 7:
  • the structure type data uses a structure identifier
  • the structure identifier is used to indicate the structure type data, that is, the structure identifier is used to determine the structure of the first model.
  • the CSI measurement feedback resources may be configured by the base station, and the base station configures F2 CSI measurement feedback resources 1 and/or P2 CSI measurement feedback resources 2 for the terminal through signaling.
  • the CSI measurement feedback resource 1 is used to carry parameter data of the first model.
  • CSI measurement feedback resource 2 is used to carry CSI.
  • the configuration of the structure of the first model and the configuration of the CSI measurement feedback resources may be sent through the same signaling, or may be sent through different signaling, and the configuration of the two information is not front-to-back order.
  • CSI measurement feedback resource 1 and CSI measurement feedback resource 2 may be delivered through one configuration signaling, or may be delivered through different configuration signaling, which is not specifically limited.
  • the structure type data has a corresponding relationship with the CSI measurement feedback resource identifier.
  • the first communication device and/or the second communication device may determine the resource corresponding to the parameter data of the first model and the resource corresponding to the second CSI according to the structure data and the correspondence between the structure data and the CSI measurement feedback resource identifier. resource.
  • the first communication device and/or the second communication device may determine the resources corresponding to the second level feedback and the resources corresponding to the third level feedback according to the structure type data and the correspondence between the structure type data and the CSI measurement feedback resource identifier.
  • Table 9 shows an example of the ninth mapping relationship.
  • the ninth mapping relationship may be one or more lines in the following table, and the ninth mapping relationship may be predefined by a protocol, or it may be notified by the second communication device to the first communication device through signaling, for example, a base station The terminal is notified through signaling, which is not specifically limited in this embodiment of the present application.
  • the base station and/or the terminal may determine CSI measurement feedback resource 1 and/or CSI measurement feedback resource 2 according to the structure of the first model.
  • the structure of the first model may have a corresponding relationship with the CSI measurement feedback resources.
  • At least one row, and/or, at least one column of the following table may be used.
  • F is a positive integer
  • P can be a positive integer. If a corresponding relationship is established between the structure identifier and the identifier of the CSI measurement feedback resource, as shown in Table 9:
  • Structure ID CSI measurement feedback resource 1 identifier CSI measurement feedback resource 2 identifier Index0 CSI measurement feedback resource 1-0 CSI measurement feedback resource 2-0 Index1 CSI measurement feedback resource 1-0 CSI measurement feedback resource 2-1 Index2 CSI measurement feedback resource 1-1 CSI measurement feedback resource 2-1 ... ... ... IndexG CSI measurement feedback resources 1-F CSI measurement feedback resource 2-P
  • the base station can send the CSI configuration information to the terminal according to the received first-level feedback, including the time-frequency resources occupied by the second-level feedback. Because the time-frequency resource overhead occupied by the second-level feedback is related to the first-level feedback, that is, the first-level feedback can determine the overhead of the second-level feedback. Therefore, the base station first receives the first-level feedback and then configures the time-frequency resources for the second-level feedback, and can determine the time-frequency resources for the second-level feedback according to the first-level feedback to achieve more reasonable resource allocation and avoid resource waste. Further, the base station may send the CSI configuration information to the terminal according to the received second-level feedback, including the time-frequency resources occupied by the third-level feedback.
  • the base station first receives the second-level feedback and then configures the time-frequency resources for the third-level feedback, and can determine the time-frequency resources for the third-level feedback according to the second-level feedback, so as to achieve more reasonable resource allocation and avoid resource waste.
  • the channel learning model when the channel learning model includes bit quantization coding and bit quantization decoding, the following manner may be used to determine the bit overhead of the CSI feedback, exemplarily the bit overhead of the second CSI.
  • the following embodiments provide a method for the first communication device and/or the second communication device to determine the bit overhead.
  • the manner of determining the bit overhead may be used as an independent embodiment, or may be combined with other embodiments, which is not limited in this application.
  • the first communication device and/or the second communication device may determine the bit overhead in at least one of the following manners.
  • One or more of the following methods for determining the bit overhead may be used alone or in combination, which is not specifically limited in this application. The following only takes the first communication device as a terminal and the second communication device as a base station as an example for description, and other situations are similar, and details are not repeated here.
  • the terminal and/or the base station determines the bit overhead based on the first model.
  • the terminal may determine the bit overhead based on the parameter class data of the first model indicated by the base station.
  • the base station may determine the bit overhead based on the parameter class data of the first model indicated by the terminal.
  • bit overhead may be determined according to the quantization rate parameter and the output dimension.
  • the parameter class data of the first model includes a quantization rate parameter and an output dimension.
  • the terminal and/or the base station determines the bit overhead according to the parameter class data of the first model.
  • bit overhead may be parameter class data of the first model. That is, the parameter class data of the first model includes bit overhead.
  • the base station includes a bit overhead when sending the parameter type data of the first model, and the terminal determines the bit overhead based on the parameter type data.
  • the terminal includes a bit overhead when sending the parameter type data of the first model, and the base station determines the bit overhead based on the parameter type data.
  • bit overhead may be structural class data of the first model. That is, the structure class data of the first model includes bit overhead.
  • the base station includes a bit overhead when sending the structure type data of the first model, and the terminal determines the bit overhead based on the structure type data.
  • the terminal includes a bit overhead when sending the structure type data of the first model, and the terminal determines the bit overhead based on the parameter structure type data.
  • the terminal determines the bit overhead based on the time-frequency resources occupied by the feedback, the modulation method and the code rate.
  • the configuration information of the CSI measurement feedback resource sent by the base station is determined according to the bit overhead.
  • the first communication device and/or the second communication device can determine the bit overhead of the CSI feedback, so as to accurately transmit and/or receive CSI, and reduce the communication between the first communication device and the second communication device. Make interaction, avoid feedback overhead and resource waste caused by signaling interaction, utilize resources reasonably, improve the accuracy of CSI feedback, and then improve communication performance.
  • the devices at both ends of the communication determine the mode of the first model, they can determine the first model according to the mode, which specifically includes determining the structural data and parameter data of the first model according to the mode.
  • structure class data and parameter class data can be sent in stages. The following describes the situation of the base station sending structure class data and parameter class data in stages. For details, please refer to FIG. 10 .
  • the following embodiments provide a method for the first communication device and/or the second communication device to determine the first model.
  • the manner of determining the first model may be used as an independent embodiment, or may be combined with other embodiments, which is not limited in this application.
  • the first communication device and/or the second communication device may determine the first model in at least one of the following manners.
  • One or more of the following methods for determining the first model can be used alone or in combination, which is not specifically limited in this application. The following only takes the first communication device as a terminal and the second communication device as a base station as an example for description, and other situations are similar, and details are not repeated here.
  • Another schematic diagram of a first model determination process in an embodiment of the present invention includes:
  • the base station determines structural data of the first model
  • the base station sends structural data to the terminal; correspondingly, the terminal receives the structural data sent by the base station.
  • the base station sends structural data to the terminal, which may also refer to the base station outputting structural data.
  • the terminal receiving the structural data sent by the base station may also refer to the terminal acquiring the structural data.
  • the base station may determine the structural data of the first model according to one or more items of scenario, uplink and downlink resources, device capability, and device location.
  • the structural data may include one or more items of Rank value, broadband amplitude, dimension, compression rate, number of feedback bits, and quantization rate, and the determined structural data may be sent to the terminal.
  • the base station determines the first feedback configuration information according to the structural data
  • the base station determines the first feedback configuration information according to the structure type data, which is similar to 903 in the foregoing embodiment. Specifically, details are not described herein again in this application.
  • the base station determines parameter type data according to the structure type data and the first feedback configuration information
  • the base station sends parameter data to the terminal; correspondingly, the terminal receives the parameter data sent by the base station.
  • the base station sends parameter-type data to the terminal, which may also mean that the base station outputs parameter-type data.
  • the terminal receiving the parameter data sent by the base station may also refer to the terminal acquiring the parameter data.
  • the terminal determines the first model according to the structural data and the parameter data.
  • an embodiment of the present application provides a communication method that can be performed by a first communication device. That is, the following method is described from the perspective of the first communication device, and the method may include at least one of S1010' to S1030':
  • the first communication device acquires structural data of the first model. Specifically, for the description of this step, reference may be made to the description in 1001 above, which is not described in detail here for brevity.
  • the first communication device acquires parameter data of the first model. Specifically, for the description of this step, reference may be made to the description in 1005 above, which is not described in detail here for brevity.
  • the first communication device determines the first model according to the structural data and the parameter data. Specifically, for the description of this step, reference may be made to the description in 1006 above, which is not described in detail here for brevity.
  • an embodiment of the present application provides a communication method that can be performed by a second communication device. That is, the following method is described from the perspective of the second communication device, and the method may include at least one of S1010" to S1050":
  • the second communication device determines the structural class data of the first model.
  • the description of this step can refer to the description in 1001 above, which is not described in detail here for brevity.
  • this step can be omitted.
  • the second communication device outputs the structural class data of the first model.
  • the description of this step can refer to the description in 1002 above, which is not described in detail here for brevity.
  • this step can be omitted.
  • the second communication device determines the first feedback configuration information according to the structural data. Specifically, for the description of this step, reference may be made to the description in 1003 above, which is not described in detail here for brevity.
  • the second communication device determines parameter type data according to the structure type data and the first feedback configuration information. Specifically, the description of this step may refer to the description in 904 above, which is not described in detail here for brevity.
  • the second communication device outputs parameter type data. Specifically, for the description of this step, reference may be made to the description in 906 above, which is not described in detail here for brevity.
  • the structural data can be sent to the terminal side, and the base station side can send the structural data to the terminal side according to The structure type data determines the first feedback configuration information, thereby determining the feedback resources of the parameter type data.
  • the base station can send the structural data and the parameter data in stages. Since the base station side does not need to determine the quantized CSI, the structural data and the parameter data are generally sent in two levels, which reduces the need for Sending structural data and parameter data at the same time occupies channel resources.
  • the base station may determine the first feedback configuration information according to the structural data, and determine the parameter data according to the first feedback configuration information together with the structural data, wherein the first feedback configuration information may be combined with the parameter data
  • the first feedback configuration information can also be sent separately from the parameter data. Sending together can avoid secondary occupation of channel resources, and is suitable for situations where downlink resources are sufficient. Separate sending can avoid single sending. Occupies more channel resources, and is suitable for situations where downlink resources are relatively insufficient.
  • the base station determines the parameter type data according to the first feedback configuration information and the determined structure type data. If the structure type data is considered to be the first level, then the parameter type data can be considered to be the second level.
  • the time-frequency resource overhead occupied by the second-level feedback is related to the first-level feedback, that is, the first-level feedback can determine the overhead of the second-level feedback. Therefore, the base station first sends the first-level feedback and then configures the time-frequency resources for the second-level feedback.
  • the time-frequency resources for the second-level feedback can be determined according to the first-level feedback, so as to achieve more reasonable resource allocation and avoid resource waste.
  • the parameter data may contain one or more kinds of parameter data
  • the parameter data may be sent in different stages in different stages.
  • the terminal sends the parameter data in stages.
  • the following embodiments provide a method for the first communication device and/or the second communication device to determine the first model.
  • the manner of determining the first model may be used as an independent embodiment, or may be combined with other embodiments, which is not limited in this application.
  • the first communication device and/or the second communication device may determine the first model in at least one of the following manners.
  • One or more of the following methods for determining the first model can be used alone or in combination, which is not specifically limited in this application.
  • the following only takes the first communication device as a terminal and the second communication device as a base station as an example for description, and other situations are similar, and details are not repeated here.
  • FIG. 11 another schematic diagram of the first model determination process in the embodiment of the present invention includes:
  • the terminal determines the second sub-data of the parameter class data
  • the terminal sends the second sub-data to the base station; correspondingly, the base station receives the second sub-data sent by the terminal.
  • the terminal sending the second sub-data to the base station may also mean that the terminal outputs the second sub-data.
  • the base station when the base station receives the second sub-data sent by the terminal, it may also mean that the base station obtains the second sub-data.
  • the first is that a part of the sub-data of the parameter class data has an influence on another part of the sub-data, such as the influence of the structure class data on the parameter class data.
  • the sub-data has the influence of inter-layer parameter class data.
  • the parameter value of the Ln th layer will affect the parameter value of the Lm th layer, where Ln and Lm are integers, and Ln is smaller than Lm.
  • the parameter value range of Ln is pn1-pn2, then it can be determined that the parameter value range of the Lm layer is also pn1-pn2, or pn3-pn4.
  • pn1, pn2, pn3, and pn4 are real numbers
  • pn3 and pn4 are determined according to pn1 and pn2.
  • the feedback can be performed by feeding back the difference value from the parameter class data of the Ln layer, thereby reducing the feedback overhead.
  • the value of the parameter class data of the Lm layer has been determined according to the parameter class data of the Ln layer. Resource waste under quantitative feedback.
  • the terminal may first feed back the parameter data of the Lnth layer, and then feed back the parameter data of the Lmth layer. In this way, the feedback overhead of parameter data can be reduced, resource utilization can be improved, and communication performance can be improved.
  • the sub-data has the influence of the parameter class data in the layer.
  • the influence of layer parameter class data is the influence between different parameters.
  • the parameters of the Lnth layer include a weight matrix and a bias matrix.
  • the value of the weight matrix will affect the value of the bias matrix.
  • the parameter value range of the weight matrix is pq1-pq2, then it can be determined that the value range of the bias parameter is also pq1-pq2, or pq3-pq4.
  • pq1, pq2, pq3, pq4 are real numbers, and pq3 and pq4 are determined according to pq1 and pq2.
  • the feedback can be performed by feeding back the difference value of the parameter class data related to the weight matrix, thereby reducing the feedback overhead.
  • the value of the parameter class data related to the bias matrix can be avoided without any preconditions.
  • the parameters are all quantified according to a wide range of resource waste under feedback.
  • the terminal may first feed back the parameter type data of the weight matrix of the Lnth layer, and then feed back the parameter type data of the bias matrix of the Lnth layer.
  • the feedback overhead of parameter data can be reduced, resource utilization can be improved, and communication performance can be improved.
  • the influence of layer parameter data is the influence between different variables of the same parameter.
  • the parameters of the Lnth layer include a weight matrix.
  • the value of the qnth variable in the weight matrix will affect the value of the qmth variable.
  • the parameter value range of the qnth variable is pqn1-pqn2, then it can be determined that the parameter value range of the qmth variable is also pqn1-pqn2, or pqn3-pqn4.
  • pqn1, pqn2, pqn3, pqn4 are real numbers, and pqn3 and pqn4 are determined according to pqn1 and pqn2.
  • the feedback can be performed by feeding back the difference between the parameter data and the parameter data of the qnth variable, thereby reducing the feedback overhead.
  • the parameter class data of the qmth variable is fed back, the value of the parameter class data of the qmth variable has been determined according to the parameter class data of the qnth variable. The waste of resources under the quantitative feedback is carried out on a large scale.
  • the terminal may first feed back the parameter data of the qnth variable of the weight matrix of the Lnth layer, and then feed back the parameter data of the qmth variable of the weight matrix of the Lnth layer.
  • the feedback overhead of parameter data can be reduced, resource utilization can be improved, and communication performance can be improved.
  • the parameter data contains many types of parameter data, and different parameter data have different time characteristics, if all parameter data are sent at the same time, it may cause unnecessary waste of channel resources, so the terminal is performing parameter data.
  • data When data is sent, it can be sent separately according to the time characteristics corresponding to each parameter data, so as to achieve more reasonable channel resource allocation.
  • the base station determines third feedback configuration information according to the second sub-data
  • the base station sends the third feedback configuration information to the terminal; correspondingly, the terminal receives the third feedback configuration information sent by the base station.
  • the base station sending the third feedback configuration information to the terminal may also mean that the base station outputs the third feedback configuration information.
  • the terminal receiving the third feedback configuration information sent by the base station may also mean that the terminal obtains the third feedback configuration information.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Software Systems (AREA)
  • Data Mining & Analysis (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Computation (AREA)
  • Medical Informatics (AREA)
  • Electromagnetism (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Quality & Reliability (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

L'invention concerne un procédé de retour de mesure d'informations d'état de canal et un appareil associé. Le procédé selon les modes de réalisation de la présente invention consiste à : déterminer un mode d'un premier modèle, le mode du premier modèle étant utilisé pour déterminer le premier modèle, et le premier modèle étant formé sur la base de données de type de structure et de données de type de paramètre ; acquérir un signal de référence d'informations d'état de canal (CSI-RS), le CSI-RS étant utilisé pour déterminer des premières informations d'état de canal (CSI) ; déterminer des secondes CSI selon le premier modèle et les premières CSI, le volume de données des secondes CSI étant inférieur au volume de données des premières CSI ; et délivrer en sortie les secondes CSI. Un mode d'un premier modèle est déterminé en fonction des conditions réelles d'un canal, et des ressources de retour d'informations d'état de canal sont agencées de manière rationnelle, ce qui permet de réduire les surdébits de retour d'informations d'état de canal.
PCT/CN2021/111700 2020-08-13 2021-08-10 Procédé de retour de mesure d'informations d'état de canal et appareil associé Ceased WO2022033456A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010811159.7 2020-08-13
CN202010811159.7A CN114079493A (zh) 2020-08-13 2020-08-13 一种信道状态信息测量反馈方法及相关装置

Publications (1)

Publication Number Publication Date
WO2022033456A1 true WO2022033456A1 (fr) 2022-02-17

Family

ID=80246913

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/111700 Ceased WO2022033456A1 (fr) 2020-08-13 2021-08-10 Procédé de retour de mesure d'informations d'état de canal et appareil associé

Country Status (2)

Country Link
CN (1) CN114079493A (fr)
WO (1) WO2022033456A1 (fr)

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2023173295A1 (fr) * 2022-03-15 2023-09-21 Nec Corporation Procédés, dispositifs et supports lisibles par ordinateur destinés à la communication
WO2023179460A1 (fr) * 2022-03-21 2023-09-28 维沃移动通信有限公司 Procédé et appareil de transmission d'informations de caractéristiques de canal, terminal, et dispositif côté réseau
US20230379067A1 (en) * 2022-05-18 2023-11-23 Rohde & Schwarz Gmbh & Co. Kg Augmented reality spectrum monitoring system
WO2024088161A1 (fr) * 2022-10-27 2024-05-02 维沃移动通信有限公司 Procédé et appareil de transmission d'informations, procédé et appareil de traitement d'informations et dispositif de communication
WO2024093997A1 (fr) * 2022-11-04 2024-05-10 维沃移动通信有限公司 Procédé et appareil de détermination d'applicabilité de modèle, et dispositif de communication
WO2024164858A1 (fr) * 2023-02-06 2024-08-15 大唐移动通信设备有限公司 Procédé d'évaluation de performances, dispositif et support d'enregistrement lisible
EP4518200A4 (fr) * 2022-05-31 2025-08-20 Zte Corp Procédé et appareil d'envoi de modèle de canal de données, et procédé et appareil d'envoi d'informations

Families Citing this family (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116827481A (zh) * 2022-03-21 2023-09-29 维沃移动通信有限公司 信道特征信息上报及恢复方法、终端和网络侧设备
CN116939636A (zh) * 2022-03-29 2023-10-24 维沃移动通信有限公司 模型有效性反馈方法、装置、终端及网络侧设备
CN116961856A (zh) * 2022-04-25 2023-10-27 展讯通信（上海）有限公司 一种数据处理方法、芯片和电子设备
CN117176216A (zh) * 2022-05-25 2023-12-05 维沃移动通信有限公司 信道状态信息csi的传输方法、装置及终端
CN117394891A (zh) * 2022-07-05 2024-01-12 中国移动通信有限公司研究院 Csi反馈的方法、装置、设备及可读存储介质
EP4420416A4 (fr) 2022-07-14 2025-01-01 ZTE Corporation Commande et gestion de modèle pour des communications sans fil
WO2024031456A1 (fr) * 2022-08-10 2024-02-15 北京小米移动软件有限公司 Procédé, appareil et dispositif de communication, support de stockage, puce, et produit-programme
CN117715072A (zh) * 2022-09-07 2024-03-15 维沃移动通信有限公司 信息传输方法、ai网络模型训练方法、装置和通信设备
CN117728869A (zh) * 2022-09-09 2024-03-19 华为技术有限公司 信道状态信息的处理方法和装置
CN117998602A (zh) * 2022-11-04 2024-05-07 华为技术有限公司 信道状态信息确定方法、装置及可读存储介质
CN119586052A (zh) * 2023-06-28 2025-03-07 北京小米移动软件有限公司 Csi反馈的确定方法、装置、设备、系统及存储介质
CN117581523A (zh) * 2023-09-26 2024-02-20 北京小米移动软件有限公司 信息传输方法及装置、通信设备、通信系统及存储介质
CN120238884A (zh) * 2023-12-29 2025-07-01 华为技术有限公司 通信方法和通信装置

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20050265436A1 (en) * 2004-06-01 2005-12-01 Samsung Electronics Co., Ltd. Method and apparatus for channel state feedback using arithmetic coding
CN109525292A (zh) * 2018-12-24 2019-03-26 东南大学 一种采用比特级优化网络的信道信息压缩反馈方法
CN109672464A (zh) * 2018-12-13 2019-04-23 西安电子科技大学 基于fcfnn的大规模mimo信道状态信息反馈方法
CN110350958A (zh) * 2019-06-13 2019-10-18 东南大学 一种基于神经网络的大规模mimo的csi多倍率压缩反馈方法
CN110708129A (zh) * 2019-08-30 2020-01-17 北京邮电大学 一种无线信道状态信息获取方法

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20050265436A1 (en) * 2004-06-01 2005-12-01 Samsung Electronics Co., Ltd. Method and apparatus for channel state feedback using arithmetic coding
CN109672464A (zh) * 2018-12-13 2019-04-23 西安电子科技大学 基于fcfnn的大规模mimo信道状态信息反馈方法
CN109525292A (zh) * 2018-12-24 2019-03-26 东南大学 一种采用比特级优化网络的信道信息压缩反馈方法
CN110350958A (zh) * 2019-06-13 2019-10-18 东南大学 一种基于神经网络的大规模mimo的csi多倍率压缩反馈方法
CN110708129A (zh) * 2019-08-30 2020-01-17 北京邮电大学 一种无线信道状态信息获取方法

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2023173295A1 (fr) * 2022-03-15 2023-09-21 Nec Corporation Procédés, dispositifs et supports lisibles par ordinateur destinés à la communication
WO2023179460A1 (fr) * 2022-03-21 2023-09-28 维沃移动通信有限公司 Procédé et appareil de transmission d'informations de caractéristiques de canal, terminal, et dispositif côté réseau
US20230379067A1 (en) * 2022-05-18 2023-11-23 Rohde & Schwarz Gmbh & Co. Kg Augmented reality spectrum monitoring system
US12155421B2 (en) * 2022-05-18 2024-11-26 Rohde & Schwarz Gmbh & Co. Kg Augmented reality spectrum monitoring system
EP4518200A4 (fr) * 2022-05-31 2025-08-20 Zte Corp Procédé et appareil d'envoi de modèle de canal de données, et procédé et appareil d'envoi d'informations
WO2024088161A1 (fr) * 2022-10-27 2024-05-02 维沃移动通信有限公司 Procédé et appareil de transmission d'informations, procédé et appareil de traitement d'informations et dispositif de communication
WO2024093997A1 (fr) * 2022-11-04 2024-05-10 维沃移动通信有限公司 Procédé et appareil de détermination d'applicabilité de modèle, et dispositif de communication
WO2024164858A1 (fr) * 2023-02-06 2024-08-15 大唐移动通信设备有限公司 Procédé d'évaluation de performances, dispositif et support d'enregistrement lisible

Also Published As

Publication number Publication date
CN114079493A (zh) 2022-02-22

Similar Documents

Publication Publication Date Title
WO2022033456A1 (fr) Procédé de retour de mesure d'informations d'état de canal et appareil associé
CN113938232A (zh) 通信的方法及通信装置
CN109873665B (zh) 数据传输的方法和设备
CN113873539B (zh) 获取神经网络的方法和装置
CN114614955B (zh) 一种传输数据的方法和装置
WO2022267633A1 (fr) Procédé et appareil de transmission d'informations
US12143983B2 (en) Scheduling method and apparatus
WO2024008004A1 (fr) Procédé et appareil de communication
CN118303110A (zh) 信息指示方法、信息处理方法和设备
WO2024046288A1 (fr) Procédé et appareil de communication
CN118118133A (zh) 一种通信方法及装置
CN118056429A (zh) 虚拟信道样本的质量评估方法和设备
CN118509014A (zh) 一种通信的方法和通信装置
WO2023092310A1 (fr) Procédé de traitement d'informations, procédé de génération de modèle et dispositifs
EP4572172A1 (fr) Procédé et appareil de communication
US20250337467A1 (en) Method and device for detecting channel variation, based on ai model in wireless communication system
CN120785389A (zh) 一种信道反馈信息传输方法及装置
WO2025067379A1 (fr) Procédé et appareil de communication, et support de stockage lisible
WO2024114686A1 (fr) Procédé d'acquisition de données d'apprentissage et appareil de communication
CN119856466A (zh) 生成样本集的方法和设备
WO2025140282A1 (fr) Procédé de communication et dispositif associé
WO2025107835A1 (fr) Procédé de communication et dispositif associé
WO2024208316A1 (fr) Procédé de rétroaction d'informations de compression et appareil
CN120956308A (zh) 通信方法、系统、装置、设备及可读存储介质
CN121126440A (zh) 用于无线通信的方法和装置

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21855508

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 21855508

Country of ref document: EP

Kind code of ref document: A1