WO2025213401A1 - Information processing method and apparatus, and communication system - Google Patents

Abstract

Provided in the embodiments of the present application are an information processing method and apparatus, and a communication system. The method comprises: a terminal device receiving CSI resource configuration information and/or CSI reporting configuration information from a network device; and on the basis of the CSI resource configuration information and/or the CSI reporting configuration information, the terminal device reporting training data used for an AI/ML model.

Description

Information processing method, device and communication system Technical Field

The embodiments of the present application relate to the field of communication technologies.

Background Art

In NR (New Radio) Rel-18 (Release 18), research was conducted on artificial intelligence/machine learning (AI/ML) for the air interface. AI/ML can be used for the following use cases: channel state information (CSI) feedback enhancement, beam management, and positioning enhancement. CSI feedback enhancement can include CSI prediction and CSI compression; beam management can include spatial beam prediction (BM case-1) and temporal beam prediction (BM case-2); and positioning enhancement can include direct positioning and AI/ML-assisted positioning. These use cases are only a preliminary selection; new use cases may be added and existing use cases may be enhanced in Rel-19 (Release 19).

With the introduction of these use cases, standardization efforts are underway to develop new protocols, processes, and signaling to support the reliable operation of AI/ML and ensure its effective benefits. These protocol-related methods and approaches are not only targeted at 5G-Advanced standards and commercial networks and devices, but can also be applied to 6G networks and devices beyond.

It should be noted that the above introduction to the technical background is merely intended to provide a clear and complete description of the technical solutions of this application and facilitate understanding by those skilled in the art. Simply because these solutions are described in the background technology section of this application, it should not be assumed that the above technical solutions are well known to those skilled in the art.

Summary of the Invention

The inventors discovered that terminal devices and/or network devices can use AI/ML models to predict CSI. AI/ML models require training data. Training data is a key factor influencing the inference performance or accuracy of AI/ML models. However, it is currently unclear how to collect this training data.

In response to at least one of the above problems, embodiments of the present application provide an information processing method, apparatus, and communication system.

According to one aspect of an embodiment of the present application, an information processing device is provided, which is configured in a terminal device, wherein the device includes: a receiving unit, which receives a channel state information (CSI) resource configuration information from a network device; information and/or channel state information (CSI) reporting configuration information; and a sending unit that reports training data for the AI/ML model based on the CSI resource configuration information and/or the CSI reporting configuration information.

According to another aspect of an embodiment of the present application, an information processing method is provided, which is applied to a terminal device, wherein the method includes: the terminal device receiving CSI resource configuration information and/or CSI reporting configuration information from a network device; and the terminal device reporting training data for an AI/ML model based on the CSI resource configuration information and/or the CSI reporting configuration information.

According to another aspect of an embodiment of the present application, an information processing device is provided, which is configured in a network device, wherein the device includes: a sending unit, which sends CSI resource configuration information and/or CSI reporting configuration information to a terminal device; and a receiving unit, which receives training data for an AI/ML model reported by the terminal device.

According to another aspect of an embodiment of the present application, an information processing method is provided, which is applied to a network device, wherein the method includes: a sending unit that sends CSI resource configuration information and/or CSI reporting configuration information to a terminal device; and a receiving unit that receives training data for an AI/ML model reported by the terminal device.

According to another aspect of an embodiment of the present application, a communication system is provided, including a network device and a terminal device, wherein the network device sends CSI resource configuration information and/or CSI reporting configuration information to the terminal device, and receives training data for an AI/ML model reported by the terminal device; the terminal device receives the CSI resource configuration information and/or CSI reporting configuration information, and reports the training data according to the CSI resource configuration information and/or CSI reporting configuration information.

One of the beneficial effects of the embodiments of the present application is that a terminal device receives CSI resource configuration information and/or CSI reporting configuration information from a network device and reports training data for an AI/ML model based on the CSI resource configuration information and/or CSI reporting configuration information. This enables the collection of training data for the AI/ML model, helping to improve the accuracy and performance of the AI/ML model.

With reference to the following description and accompanying drawings, specific embodiments of the present application are disclosed in detail, indicating the manner in which the principles of the present application can be employed. It should be understood that the embodiments of the present application are not limited in scope. Within the spirit and scope of the appended claims, the embodiments of the present application include many variations, modifications and equivalents.

Features described and/or illustrated with respect to one embodiment may be used in the same or similar manner in one or more other embodiments, combined with features in other embodiments, or substituted for features in other embodiments.

It should be emphasized that the term "include/comprising" when used herein refers to the presence of features, integers, steps or components, but does not exclude the presence or addition of one or more other features, integers, steps or components.

BRIEF DESCRIPTION OF THE DRAWINGS

The elements and features described in one figure or one embodiment of the present application can be combined with the elements and features shown in one or more other figures or embodiments. In addition, in the accompanying drawings, similar reference numerals represent corresponding parts in several figures and can be used to indicate corresponding parts used in more than one embodiment.

FIG1 is a schematic diagram of a communication system according to an embodiment of the present application;

FIG2 is a schematic diagram of an information processing method according to an embodiment of the present application;

FIG3 is another schematic diagram of the information processing method according to an embodiment of the present application;

FIG4 is a schematic diagram of an information processing device according to an embodiment of the present application;

FIG5 is another schematic diagram of the information processing device according to an embodiment of the present application;

FIG6 is a schematic diagram of a network device according to an embodiment of the present application;

FIG7 is a schematic diagram of a terminal device according to an embodiment of the present application.

DETAILED DESCRIPTION

The foregoing and other features of the present application will become apparent from the following description with reference to the accompanying drawings. In the description and drawings, specific embodiments of the present application are disclosed, which illustrate some embodiments in which the principles of the present application can be employed. It should be understood that the present application is not limited to the described embodiments. On the contrary, the present application includes all modifications, variations, and equivalents falling within the scope of the appended claims. Various embodiments of the present application are described below with reference to the accompanying drawings. These embodiments are merely illustrative and are not intended to limit the present application.

In the embodiments of the present application, the terms "first", "second", etc. are used to distinguish different elements from the name, but do not indicate the spatial arrangement or temporal order of these elements, and these elements should not be limited by these terms. The term "and/or" includes any one and all combinations of one or more of the associated listed terms. The terms "comprising", "including", "having", etc. refer to the presence of the stated features, elements, components or components, but do not exclude the presence or addition of one or more other features, elements, components or components.

In the embodiments of this application, the singular forms "a,""the," etc. include plural forms and should be broadly understood to mean "a" or "a type" rather than being limited to "one." Furthermore, the term "said" should be understood to include both singular and plural forms, unless the context clearly indicates otherwise. Furthermore, the term "according to" should be understood to mean "at least in part based on...", and the term "based on" should be understood to mean "at least in part based on...", unless the context clearly indicates otherwise.

In the embodiments of the present application, the term "communication network" or "wireless communication network" may refer to a network that complies with any of the following communication standards, such as New Radio (NR), Long Term Evolution (LTE), Enhanced Long Term Evolution (LTE-A, LTE-Advanced), Wideband Code Division Multiple Access (WCDMA), High-Speed Packet Access (HSPA), etc.

Furthermore, communication between devices in the communication system may be carried out according to communication protocols of any stage, such as but not limited to the following communication protocols: 1G (generation), 2G, 2.5G, 2.75G, 3G, 4G, 4.5G and 5G, New Radio (NR), 6G and future communications, etc., and/or other communication protocols currently known or to be developed in the future.

In the embodiments of the present application, the term "network device" refers to, for example, a device in a communication system that connects a terminal device to a communication network and provides services for the terminal device. Network devices may include, but are not limited to, base stations (BS), access points (AP), transmission reception points (TRP), broadcast transmitters, mobile management entities (MME), gateways, servers, radio network controllers (RNC), base station controllers (BSC), and the like.

Among them, base stations may include but are not limited to: NodeB (NodeB or NB), evolved NodeB (eNodeB or eNB), 5G base station (gNB), 6G base station and future base stations, etc., and may also include remote radio heads (RRH, Remote Radio Head), remote radio units (RRU, Remote Radio Unit), relays or low-power nodes (such as femto, pico, etc.). The term "base station" may include some or all of their functions. Each base station can provide communication coverage for a specific geographical area. The term "cell" can refer to a base station and/or its coverage area, depending on the context in which the term is used.

In the embodiments of the present application, the term "user equipment" (UE) or "terminal equipment" (TE) refers to, for example, a device that accesses a communication network through a network device and receives network services. A user equipment may be fixed or mobile and may also be referred to as a mobile station (MS), a terminal, a user, a subscriber station (SS), an access terminal (AT), a station, a mobile terminal (MT), and so on.

The terminal device may include, but is not limited to, the following devices: cellular phones, personal digital assistants (PDAs), wireless modems, wireless communication devices, handheld devices, machine-type communication devices, laptop computers, cordless phones, smartphones, smart watches, digital cameras, etc.

For another example, in scenarios such as the Internet of Things (IoT), the user equipment may also be a machine or device for monitoring or measurement, including but not limited to: machine type communication (MTC) terminals, vehicle-mounted communication terminals, device-to-device (D2D) terminals, machine-to-machine (M2M) terminals, terminals supporting sidelink communication, and the like.

In addition, the term "network side" or "network device side" refers to one side of the network, which can be a base station or one or more network devices as described above. The term "user side" or "terminal side" or "terminal device side" refers to the user or terminal side, which can be a UE or one or more terminal devices as described above. Unless otherwise specified herein, "device" can refer to either network equipment or terminal equipment.

In order to avoid confusion, the terms "uplink control signal" and "uplink control information (UCI)" or "physical uplink control channel (PUCCH)" are interchangeable, and the terms "uplink data signal" and "uplink data information" or "physical uplink shared channel (PUSCH)" are interchangeable.

The terms "downlink control signal" and "downlink control information (DCI)" or "physical downlink control channel (PDCCH)" are interchangeable, and the terms "downlink data signal" and "downlink data information" or "physical downlink shared channel (PDSCH)" are interchangeable.

In addition, the uplink signal may include an uplink data signal and/or an uplink control signal and/or a PRACH and/or an SRS (sounding reference signal), etc., which may also be referred to as an uplink transmission (UL transmission) or an uplink information or an uplink channel. Sending/receiving an uplink transmission on an uplink resource may be understood as sending/receiving the uplink transmission using the uplink resource. The downlink signal may include a downlink data signal and/or a downlink control signal and/or a synchronization signal (SS, such as PSS/SSS) and/or a broadcast channel (PBCH) and/or an SSB (SS/PBCH block, including PSS, SSS and PBCH and its DMRS) and/or a CSI-RS, etc., which may also be referred to as a downlink transmission (DL transmission) or a downlink information or a downlink channel. Sending/receiving a downlink transmission on a downlink resource may be understood as sending/receiving the downlink transmission using the downlink resource.

In the embodiment of the present application, the high-level signaling may be, for example, radio resource control (RRC) signaling; RRC signaling may include, for example, an RRC message, for example, a broadcast/public RRC message/signaling (e.g., a master information block (MIB), system information (SI), a dedicated RRC message/signaling; or an RRC Information element (RRC information element, RRC IE); or an information field included in an RRC message or RRC information element (or an information field included in an information field). Higher-layer signaling may also be, for example, Medium Access Control (MAC) signaling; or called a MAC control element (MAC control element, MAC CE). However, the present application is not limited thereto.

In the embodiments of the present application, "at least one" and "one or more than one" can be interchanged, "multiple" and "more than one" can be interchanged, and "multiple" means at least two, or two or more than two.

In the embodiments of the present application, predefined means specified in the protocol or determined according to the rules specified in the protocol, and no additional configuration is required. Configuration/indication refers to direct or indirect configuration/indication by the network device through high-layer signaling and/or physical layer signaling. Configuration/indication can be achieved by introducing high-layer parameters in high-layer signaling, and high-layer parameters refer to information fields (fields) and/or information elements/information units/information elements (IEs) in high-layer signaling. Physical layer signaling refers to, for example, control information (DCI) carried by the physical downlink control channel or control information carried by the sequence, but is not limited thereto.

For ease of description, the following description will be made using a base station as an example of an access network device. In the following description, "if ...", "under ..." and "when ..." can be used interchangeably without causing confusion.

The following describes the scenarios of the embodiments of the present application through examples, but the present application is not limited thereto.

FIG1 is a schematic diagram of a communication system according to an embodiment of the present application, schematically illustrating a situation taking a terminal device and a network device as an example. As shown in FIG1 , a communication system 100 may include a network device 101, a terminal device 102, and a terminal device 103. For simplicity, FIG1 illustrates only two terminal devices and one network device as an example, but the embodiments of the present application are not limited thereto.

In the embodiment of the present application, existing services or future services can be transmitted between the network device 101, the terminal device 102, and the terminal device 103. For example, these services may include, but are not limited to, enhanced mobile broadband (eMBB), massive machine type communication (mMTC), ultra-reliable and low-latency communication (URLLC), and communications related to reduced-capability terminal devices, etc.

Among them, terminal devices 102 and 103 can be in RRC_IDLE state, RRC_INACTIVE state or RRC_CONNECTED state, and terminal devices 102 and 103 can also communicate with network device 101. For example, taking terminal device 102 as an example, terminal device 102 can send data to network device 101, or can retransmit data. Network device 101 can send a paging message to terminal device 102, or send data to terminal device 102, and terminal device 102 receives the data sent by network device 101. In addition, different terminal devices Communication can also be performed between them, for example, data can be exchanged between terminal device 102 and terminal device 103.

It is worth noting that FIG1 shows that terminal device 102 and terminal device 103 are both within the coverage range of network device 101, but the present application is not limited thereto. Terminal device 102 and terminal device 103 may both be outside the coverage range of network device 101, or one of terminal device 102 and terminal device 103 may be within the coverage range of network device 101 and the other may be outside the coverage range of network device 101.

In embodiments of the present application, one or more AI/ML models may be configured and run in a network device and/or a terminal device. The AI/ML models may be used for various signal processing functions in wireless communications, such as CSI prediction, CSI compression, beamforming, positioning management, and the like; however, the present application is not limited thereto.

Regarding the CSI prediction sub-use case, the relevant content is shown in Tables 1 and 2:

Table 1

Table 2

To achieve good CSI prediction accuracy, AI/ML-based CSI prediction requires consistency between AI/ML model training and inference. This means that during the training data collection phase, the device and network configurations, auxiliary information, or additional conditions must be consistent with those used during the inference phase. Inconsistencies between the corresponding information in the training and inference phases can cause input distribution drift in the AI/ML model, preventing it from operating optimally.

On the other hand, compared to universal AI/ML models, local AI/ML models often achieve better performance and lower complexity. To train local AI/ML models or achieve local gains, we must classify the training data. Using different types of training data to train AI/ML models separately helps achieve local gains.

Therefore, for the collection of training data, in addition to the data related to the model input and model output (label/ground truth), we also need to collect configuration information, auxiliary information, and additional conditions on the network side and the terminal side.

In the embodiments of the present application, AI/ML may also be referred to as AI/ML model, AI/ML method, AI/ML unit, AI/ML functionality, or similar names of AI/ML element.

In the embodiments of the present application, the term "true value" can be replaced by equivalent names such as "label", "ground truth", "actual value", etc.

Embodiments of the first aspect

The present application embodiment provides an information processing method, which is described from the perspective of a terminal device. FIG2 is a schematic diagram of the information processing method of the present application embodiment. As shown in FIG2 , the method includes:

201: The terminal device receives CSI resource configuration information and/or CSI reporting configuration information from the network device; and

202: The terminal device reports the CSI resource configuration information and/or the CSI reporting configuration information. Training data for AI/ML models.

It is worth noting that FIG2 above is merely a schematic illustration of an embodiment of the present application, and the present application is not limited thereto. For example, the execution order of the various operations may be appropriately adjusted, and other operations may be added or some operations may be reduced. Those skilled in the art may make appropriate modifications based on the above description, and are not limited to the description of FIG2 above.

According to the above embodiment, the terminal device receives CSI resource configuration information and/or CSI reporting configuration information from the network device and reports training data for the AI/ML model based on the CSI resource configuration information and/or CSI reporting configuration information. This enables the collection of training data for the AI/ML model, helping to improve the accuracy and performance of the AI/ML model.

In some embodiments, the UE may use a model (AI/ML model) based on artificial intelligence (AI) and/or machine learning (ML) to predict CSI. For example, the UE may use the AI/ML model to predict one or more CSI-related information (output information, which may be referred to as CSI) at a future moment based on one or more CSI-related information (input information, which may be referred to as CSI). As a result, at one or more moments in the future, the network device does not need to send CSI-RS to the terminal device, thereby reducing signaling overhead and contributing to energy saving on the network side; or, the terminal device does not need to measure CSI-RS or report CSI at one or more moments in the future, thereby reducing signaling overhead and contributing to energy saving on the terminal side; or, the network device uses CSI-related information at one or more moments in the future to perform downlink PDSCH transmission, thereby reducing performance loss caused by aging of the feedback CSI.

In some embodiments, the UE may use an AI/ML model to predict L second CSI-related information based on K first CSI-related information, where K and L are integers greater than or equal to 1. For example, K=4, L=2, and the UE predicts second CSI-related information at time points t+2Δ and t+4Δ based on the first CSI-related information at time points t-3Δ, t-2Δ, t-Δ, and t, where Δ is greater than 0.

The first CSI-related information and the second CSI-related information may be a channel matrix and/or an eigenvector.

The channel matrix is, for example, a matrix used to describe the characteristics of a transmission channel. The channel matrix can typically be used to describe the attenuation and/or distortion of a transmitted signal during transmission in the channel. The channel matrix can be obtained by transmitting a reference signal and performing channel estimation.

An eigenvector is, for example, a vector obtained by performing matrix decomposition of a channel matrix, such as SVD (singular value decomposition) or EVD (eigenvalue decomposition). In some embodiments, multiple layers of precoding vectors can be combined into a precoding matrix. In the embodiments of this application, the terms "eigenvector" and "precoding vector" are interchangeable.

The present application is not limited thereto, and the first CSI-related information (input information) and the second CSI-related information (output information) may also be other forms of information capable of indicating the channel state.

In some embodiments, the AI/ML model may predict L channel matrices based on K channel matrices; alternatively, the AI/ML model may predict L eigenvectors based on K eigenvectors.

However, the present application is not limited thereto. For example, the AI/ML model may also predict L eigenvectors based on K channel matrices, or, alternatively, predict L channel matrices based on K eigenvectors. For another example, the input and/or output of the AI/ML model may also be a combination of a channel matrix and an eigenvector. For example, the AI/ML model may predict the channel matrix based on the channel matrix and the eigenvector, or, predict the eigenvector based on the channel matrix and the eigenvector, or, predict the channel matrix and the eigenvector based on the channel matrix, or, predict the channel matrix and the eigenvector based on the eigenvector, or, predict the channel matrix and the eigenvector based on the channel matrix and the eigenvector, and so on.

In some embodiments, the training data of the AI/ML model may include at least one of the following: input information of the model; a true value corresponding to the output information of the model; or first information of the terminal device.

The input information of the model may include a channel matrix and/or an eigenvector. The input information may be generated based on a CSI-RS measurement result at a first moment (also referred to as a measurement moment).

For example, in 202, the UE may measure one or more CSI-RSs at a first time (e.g., the aforementioned time t-3Δ, t-2Δ, t-Δ, t) based on the received CSI resource configuration information and/or CSI reporting configuration information, generate a channel matrix and/or eigenvector based on the measurement results, and use the channel matrix and/or eigenvector as input information for the model. For ease of description, the CSI-RS corresponding to the input information of the model is referred to as a first CSI-RS, and the resource corresponding to the first CSI-RS is referred to as a first resource.

The true value corresponding to the model's output information may include a channel matrix and/or an eigenvector. In embodiments of the present application, the model's output information is output information generated based on the model's input information during the model inference process; the true value corresponding to the model's output information is the model's output information under ideal conditions. One of the purposes of training an AL/ML model is to make the model's output information as close as possible to the true value corresponding to the output information.

The true value corresponding to the output information may be generated according to the measurement result of the CSI-RS at the second moment (also referred to as the prediction moment).

For example, in 202, the UE may measure the CSI-RS at one or more second moments (e.g., the aforementioned moments t+2Δ, t+4Δ) according to the received CSI resource configuration information and/or CSI reporting configuration information, generate a channel matrix and/or eigenvector according to the measurement results, and use the channel matrix and/or eigenvector as the model. For ease of description, the CSI-RS corresponding to the output information of the model is referred to as the second CSI-RS, and the resource corresponding to the second CSI-RS is referred to as the second resource.

In some embodiments, the first information of the terminal device can also be called auxiliary information (assistance information) or additional condition (additional condition) of the terminal device.

The first information of the terminal device may include at least one of the following: antenna configuration information of the terminal device, moving speed information of the terminal device, or expected time information of the terminal device.

Since the true values corresponding to the model's input and output information are usually related to the first information, including this first information in the training data can serve as a marker for the training data (also known as a "mark," "index," or similar names), thereby ensuring consistency between model training and model inference. That is, the first information in the model training phase can be kept consistent with the first information in the model inference phase. This can improve the prediction accuracy of the AI/ML model.

On the other hand, by including the first information in the training data, the real values corresponding to the model's input information and/or output information can be classified based on the first information. Thus, AI/ML models corresponding to the respective classifications can be trained based on the classified training data, which helps improve the accuracy of the AI/ML models, reduce the complexity of model training, and achieve better performance.

During the model inference phase, network devices and/or terminal devices can perform LCM (Life-Cycle Management) operations such as model selection/model activation/model deactivation/model switching/rollback based on classification labels.

For example, when the current first information and/or second information is consistent with the classification label of the first model (or the first information and/or second information corresponding to a certain model), the first model should be selected or activated; or, when the current first information and/or second information is inconsistent with the classification label of the second model (or the first information and/or second information corresponding to a certain model), the second model should be deactivated.

For another example, when the current first information and/or second information is inconsistent with the classification label of the current first model (or the first information and/or second information corresponding to a certain model), and the current first information and/or second information is consistent with the classification label of the second model (or the first information and/or second information corresponding to a certain model), the first model can be switched to the second model.

For another example, when there is no model consistent with the current first information and/or second information, it is possible to fall back from the AI/ML model/function to a non-AI/ML method (ie, a traditional method) for corresponding processing.

In some embodiments, the antenna configuration information of the terminal device may include at least one of the following: horizontal direction The number of panels and the number of vertical panels; the horizontal panel spacing and the vertical panel spacing; the number of horizontal antennas within a panel and the number of vertical antennas within a panel; the horizontal antenna spacing within a panel and the vertical antenna spacing within a panel; the mapping relationship between antennas and TxRUs; or the antenna polarization direction. This application is not limited to this, and the antenna configuration information of the terminal device may also include other information.

In some embodiments, the time information expected by the terminal device may include at least one of the following: the first time information for performing CSI-RS measurement, the second time information for performing CSI prediction, or the interval between the first time for performing CSI-RS measurement and the second time for performing CSI prediction.

The first moment information may include the number and/or interval of CSI-RS measurements. The number of CSI-RS measurements may refer to the number of channel matrices and/or eigenvectors included in the input information during one inference process. The interval for CSI-RS measurements may refer to the time interval between the CSI-RSs corresponding to two adjacent channel matrices and/or eigenvectors in the input information. Taking the channel matrices at the four moments t-3△, t-2△, t-△, and t as an example, the number of CSI-RS measurements is K=4, and the interval for CSI-RS measurements is △.

The second moment information may include the number and/or interval of predicted CSI. The number of predicted CSI may refer to the number of channel matrices and/or eigenvectors included in the output information during one inference process. The interval of predicted CSI may refer to the time interval of CSI-RS corresponding to two adjacent channel matrices and/or eigenvectors in the output information. Taking the channel matrices at the four moments t-3△, t-2△, t-△, t as an example, the number of predicted CSI is L=2, and the interval for CSI-RS measurement is 2△.

The number of times the CSI-RS measurements are performed may be the same as or different from the number of predicted CSIs, and the interval of the CSI-RS measurements may be the same as or different from the interval of the predicted CSIs.

In some embodiments, when there are multiple first moments for performing CSI-RS measurement and/or second moments for performing CSI prediction, the interval between the first moment and the second moment can be expressed as the interval between the last first moment and the first second moment. Taking the channel matrices at moments t-3Δ, t-2Δ, t-Δ, and t as an example, predicting the channel matrices at moments t+2Δ and t+4Δ, the interval between the first moment and the second moment is 2Δ. The present application is not limited to this, and the interval can be expressed as the interval between any first moment and any second moment.

In some embodiments, when classifying and/or labeling the training data, the second information of the network device may also be referred to, or the first information of the terminal device and the second information of the network device may be referred to. The second information of the network device may also be referred to as assistance information or additional condition of the network device.

In some embodiments, the second information includes at least one of the following: antenna configuration information of the network device, scenario information of the network device, reference signal period, cell/site identification, carrier frequency, frequency domain granularity, or subcarrier spacing.

In some embodiments, the antenna configuration information of the network device may include at least one of the following: the number of horizontal panels and the number of vertical panels; the horizontal panel spacing and the vertical panel spacing; the number of horizontal antennas within the panel and the number of vertical antennas within the panel; the horizontal antenna spacing within the panel and the vertical antenna spacing within the panel; the mapping relationship between the antenna and the TxRU; or the antenna polarization direction.

In some embodiments, the scene information of the network device may include: indoor or outdoor, and/or line-of-sight or non-line-of-sight.

In some embodiments, the frequency domain granularity may include the number of PRBs (Physical Resource Blocks) in the subband. The present application is not limited thereto, and the frequency domain granularity may also include other information.

In some embodiments, the classification and/or labeling of training data may be performed on the terminal side or on the network side. For example, the UE includes first information in the reported training data, and the network device may classify and/or label the training data based on the first information of the UE and/or the second information of the network device. For another example, the UE receives second information sent by the network device, and the UE may classify and/or label the training data based on the first information and/or the second information, and report the classified and/or labeled training data.

The following is an exemplary description of the reference signal and reporting configuration used for training data collection.

In some embodiments, the CSI resource configuration information may be used to configure one or more CSI-RS resources.

The period or time interval of the CSI-RS resources can be configured based on information about the AI/ML model. For example, the period or time interval of the CSI-RS resources can be configured to be the same as the period or time interval of the training data used by the AI/ML model during training. This ensures the validity and reliability of the collected training data. Alternatively, the period or time interval of the CSI-RS resources can be configured to be the same as the period or time interval of the input data used by the AI/ML model during inference. This ensures consistency between AI/ML model training and inference.

For example, the training data used by the AI/ML model during training may have equal periods or evenly spaced intervals in the time domain. In this case, the CSI-RS resources include at least one of the following: periodic CSI-RS resources, semi-persistent CSI-RS resources, or evenly spaced intervals in the time domain. Aperiodic CSI-RS (aperiodic CSI-RS burst) resources.

However, the present application is not limited thereto, and the periodicity or time domain interval of the CSI-RS resources may be smaller than the periodicity or time interval of the training data used in the training process or the input data used in the inference process. In other words, part of the CSI-RS resources configured by the CSI resource configuration information is used for collecting training data. For example, the periodicity or time interval of the training data or input data may be an integer multiple of the periodicity or time domain interval of the CSI-RS resources. Thus, during the training data collection phase, the UE may perform measurements on part of the CSI-RS resources.

In some embodiments, the CSI-RS resources may be configured according to at least one of the following: a scenario of the network device, a moving speed of the terminal device, or a configuration desired by the terminal device.

The number of moments and/or intervals of CSI-RS resources can be configured based on at least one of the above information. For example, when the terminal device is moving at a low speed, the time domain correlation of CSI is large. In this case, the AI/ML model can use fewer first CSI-related information or information with larger time domain intervals to predict second CSI-related information; and/or the AI/ML model can predict more second CSI-related information or information with larger time domain intervals based on the first CSI-related information. Therefore, when collecting training data, less input information and/or input information with larger time domain intervals can be collected, and/or more output information and/or output information with larger time domain intervals and/or output information with larger time domain intervals from the input information can be collected. That is, the number of measurement moments for CSI-RS resources can be reduced, and/or the measurement moment intervals for CSI-RS resources can be increased, and/or the number of prediction moments for CSI-RS resources can be increased, and/or the prediction moment intervals for CSI-RS resources can be increased, and/or the interval between the measurement moments for CSI-RS resources and the prediction moments for CSI-RS resources can be increased.

On the contrary, when the terminal device moves faster, the channel correlation is poor. In this case, the AI/ML model needs to use more first CSI-related information or smaller time domain intervals to predict the second CSI-related information; and/or the AI/ML model can predict fewer second CSI-related information or smaller time domain intervals based on the first CSI-related information. Thus, the number of measurement moments of CSI-RS resources can be increased, and/or the measurement moment interval of CSI-RS resources can be reduced, and/or the number of prediction moments of CSI-RS resources can be reduced, and/or the prediction moment interval of CSI-RS resources can be reduced, and/or the interval between the measurement moment of CSI-RS resources and the prediction moment of CSI-RS resources can be reduced.

Similarly, the number of time instants and/or intervals of CSI-RS resources can also be configured based on the scenario of the network device (e.g., indoor or outdoor, and/or line-of-sight or non-line-of-sight). For example, when the scenario is indoors or line-of-sight, the time domain correlation of CSI is relatively good, while when the scenario is outdoor or non-line-of-sight, the time domain correlation of CSI is relatively poor.

In addition, the number of moments and/or intervals of CSI-RS resources may also be configured according to the configuration desired by the terminal device. The configuration expected by the terminal device includes, for example, the time information expected by the terminal device. For example, the number of time moments and/or the interval of the CSI-RS resources are configured to be the same as the time information expected by the terminal device. Alternatively, the number of time moments of the CSI-RS resources is configured to be a value greater than the number of time moments expected by the terminal device, and/or the interval of the CSI-RS resources is configured to be a value smaller than the interval expected by the terminal device. The terminal device can select a portion of the CSI-RS resources from the configured CSI-RS resources for collecting training data.

In some embodiments, the CSI-RS resources used for reporting training data can also be used for acquiring and/or reporting channel state information (CSI), i.e., legacy CSI acquisition and/or reporting. In other words, the reference signals used for training data collection can be associated with layer 1 CSI reporting. For example, layer 1 CSI reporting can include PMI/RI/CQI/RSRP, etc. This means that the reference signals used for training data collection can be used for both legacy CSI measurement and reporting.

In some embodiments, the CSI-RS resources used for reporting training data can also be used for model performance monitoring. That is, the CSI-RS resources can be used simultaneously for AI/ML model performance monitoring and training data reporting. In other words, the reference signal used for training data collection can be used for AI/ML model performance monitoring.

In some embodiments, the UE may report training data based on the results of the performance monitoring of the AI/ML model. For example, when the performance monitoring result of the model is lower than a first threshold, the training data is reported; and/or, when the performance monitoring result of the model is higher than or equal to the first threshold, the training data is not reported. Thus, the reported training data corresponds to the situation where the performance of the AI/ML model is poor, that is, the situation where training needs to be focused, so that training data can be collected in a targeted manner, which helps to improve the performance of the AI/ML model through training.

The present application is not limited thereto. When the performance monitoring result of the model is higher than the second threshold, the training data is reported; and/or when the performance monitoring result of the model is lower than or equal to the second threshold, the training data is not reported. Thus, the reported training data corresponds to the situation where the performance of the AI/ML model is better. By training the model based on the above training data, the performance of the AI/ML model is improved.

In some embodiments, the second threshold may be a value greater than the first threshold.

In some embodiments, the first threshold and/or the second threshold may be predefined or network configured.

In some embodiments, the CSI-RS resources may include a first resource corresponding to a first CSI-RS and a second resource corresponding to a second CSI-RS. As described above, the measurement result of the first CSI-RS is used to generate input information of the AI/ML model, and the measurement result of the second CSI-RS is used to generate a true value corresponding to the output information of the AI/ML model.

The first resource and the second resource may be configured in various ways.

For example, the first resource and the second resource may be configured separately. For example, the network device may indicate which CSI-RS resources are the first resource and which CSI-RS resources are the second resource.

The first resource is associated with the second resource. For example, the network device may associate the first resource with the second resource through signaling, such as by performing resource association through a new IE during RRC configuration. Alternatively, the terminal device may associate the first resource with the second resource based on desired time information.

For another example, the first resource and the second resource may be jointly configured. For example, the network device may not explicitly distinguish between the first resource and the second resource. For example, the network device may configure CSI-RS resources for a period, and the terminal device may select the first resource and the second resource from the CSI-RS resources of the period based on desired time information.

In some embodiments, the measurement results of the first CSI-RS and the second CSI-RS may correspond to the same receive beam (Rx beam). This eliminates the impact of different receive beams on the measurement results, reduces the variables in the training data, and simplifies the training process. The present application is not limited to this, and the measurement results of the first CSI-RS and the second CSI-RS may also correspond to different receive beams.

In some embodiments, the CSI reporting configuration information and/or the CSI resource configuration information may include first indication information for instructing the collection and/or reporting of training data. Thus, after receiving the first indication information, the UE may collect and/or report the training data.

In some embodiments, the first indication information may be indicated explicitly, for example, by the appearance or absence of the first indication information, or a numerical value to indicate whether to report the training data. For example, when the first indication information appears in the CSI reporting configuration information and/or the CSI resource configuration information, the UE is instructed to report the training data; when the first indication information is absent in the CSI reporting configuration information and/or the CSI resource configuration information, the UE is instructed not to report the training data; and vice versa. For another example, the indication is made through 1-bit first indication information: when the value of the 1-bit information is 1, the UE is instructed to report the training data; when the value of the 1-bit information is 0, the UE is instructed not to report the training data; and vice versa. The present application is not limited to this, and the first indication information may also be indicated implicitly. For example, when the reporting information related to the training data or the resources related to the training data are configured, the training data is reported by default, otherwise the training data is not reported.

In some embodiments, training data may be reported in various ways.

In some embodiments, the training data may be reported via non-layer 1 signaling, such as layer 3 signaling, etc. In this case, the first indication information may be carried in radio resource control (RRC) signaling.

In some embodiments, the training data may be reported via layer 1 signaling, such as uplink control information (UCI). In this case, the first indication information may be carried on at least one of the following: radio resource control (RC) (RRC) signaling, media access control layer control element (MAC CE), or downlink control information (DCI).

In some embodiments, the training data may be reported once, or divided into multiple reports.

For example, the corresponding true values of input information and output information can be reported jointly, that is, one report includes both input information and output information.

For example, in the joint reporting, the true values corresponding to the input information and the output information can be reported jointly as a whole. That is, the channel matrix/eigenvector is reported once after the reference signal is sent.

Taking the channel matrices at time points t+2△ and t+4△ as an example, the channel matrices at time points t-3△, t-2△, t-△, and t are predicted. The channel matrices obtained by measuring the CSI-RS at time points t-3△, t-2△, t-△, and t, and the channel matrices obtained by measuring the CSI-RS at time points t+2△ and t+4△ can be reported to the network device at one time.

The present application is not limited thereto, and in the joint reporting, the input information and the output information may also be reported multiple times.

For another example, the corresponding true values of input information and output information can be reported independently. That is, in one report, the input information is included but the output information is not included; or the output information is included but the input information is not included.

For example, when input information is reported independently, the input information may be reported as a whole; when output information is reported independently, the true value corresponding to the output information may be reported as a whole.

Taking the channel matrices at time points t+2△ and t+4△ as an example, which are predicted based on the channel matrices at time points t-3△, t-2△, t-△, and t, the channel matrices obtained by measuring the CSI-RS at time points t-3△, t-2△, t-△, and t can be reported once; and the channel matrices obtained by measuring the CSI-RS at time points t+2△ and t+4△ can be reported again.

The present application is not limited thereto, and the input information may also be reported multiple times. For example, the input information may include multiple CSI-RS measurement results at a first moment (i.e., multiple channel matrices and/or eigenvectors). In this case, these measurement results may be reported multiple times.

For example, using the channel matrices at time points t-3Δ, t-2Δ, t-Δ, and t- to predict the channel matrices at time points t+2Δ and t+4Δ, the channel matrices at time points t-3Δ, t-2Δ, t-Δ, and t- can be reported four times. For example, after measuring the CSI-RS at time point t-3Δ, the UE reports the channel matrix corresponding to that time point, i.e., the channel matrix at time point t-3Δ. After measuring the CSI-RS at time point t-2Δ, the UE reports the channel matrix corresponding to that time point, i.e., the channel matrix at time point t-2Δ. And so on. In other words, the channel matrix/eigenvector can be reported multiple times after each time point during reference signal transmission.

For output information, multiple reports can also be made. For example, the output information may include multiple second moments. The measurement results of the CSI-RS (ie, multiple channel matrices and/or eigenvectors) may be reported in multiple times in this case.

Taking the channel matrices at time points t-3Δ, t-2Δ, t-Δ, and t as an example, the channel matrices at time points t+2Δ and t+4Δ can be predicted based on the channel matrices at time points t-3Δ, t-2Δ, t-Δ, and t-Δ. The channel matrices at time points t+2Δ and t+4Δ can be reported in two batches. For example, after completing the CSI-RS measurement at time point t+2Δ, the UE reports the channel matrix corresponding to that time point, i.e., the channel matrix at t+2Δ. After completing the CSI-RS measurement at time point t+4Δ, the UE reports the channel matrix corresponding to that time point, i.e., the channel matrix at t+4Δ.

The above is an exemplary description of the reporting method of the input information and the output information. When the training data includes the first information, it can also be reported once or multiple times.

For example, the input information, output information and first information may be reported jointly; or, the input information, output information and first information may be reported independently; or, any two of the input information, output information and first information may be reported jointly, and the other may be reported independently.

In some embodiments, when multiple channel matrices and/or eigenvectors are reported at once, the complete contents of each channel matrix and/or eigenvector may be reported in the report. For example, the complete contents of each channel matrix and/or eigenvector may be concatenated and the concatenated result reported.

The present application is not limited thereto. When multiple channel matrices and/or eigenvectors are reported at one time, the multiple channel matrices and/or eigenvectors may be appropriately compressed to reduce signaling overhead.

For example, the complete content of at least one channel matrix and/or eigenvector is reported in this one report, and the channel matrix and/or eigenvector is used as a reference channel matrix and/or reference eigenvector. For other channel matrices and/or eigenvectors, their differences with the reference channel matrix and/or reference eigenvector can be reported.

For example, use AI/ML models to compress reported training data, etc.

The above embodiments are merely exemplary of the present invention, but the present invention is not limited thereto. Appropriate modifications may be made based on the above embodiments. For example, the above embodiments may be used alone, or one or more of the above embodiments may be combined.

According to the above embodiment, the terminal device receives CSI resource configuration information and/or CSI reporting configuration information from the network device and reports training data for the AI/ML model based on the CSI resource configuration information and/or CSI reporting configuration information. This enables the collection of training data for the AI/ML model, helping to improve the accuracy and performance of the AI/ML model.

Embodiments of the second aspect

The embodiment of the present application provides an information processing method, which is described from the perspective of a network device. The embodiment of the second aspect can be combined with the embodiment of the first aspect, or implemented separately, and the same contents as the embodiment of the first aspect will not be repeated.

FIG3 is another schematic diagram of the information processing method according to an embodiment of the present application. As shown in FIG3 , the method includes:

301, the network device sends CSI resource configuration information and/or CSI reporting configuration information to the terminal device; and

302. The network device receives training data for the AI/ML model reported by the terminal device.

It is worth noting that FIG3 above is merely a schematic illustration of an embodiment of the present application, and the present application is not limited thereto. For example, the execution order of the various operations may be appropriately adjusted, and other operations may be added or some operations may be reduced. Those skilled in the art may make appropriate modifications based on the above description, and are not limited to the description of FIG3 above.

In some embodiments, the training data includes at least one of the following: input information of the model; a true value corresponding to output information of the model; or first information of the terminal device.

In some embodiments, the input information includes a channel matrix and/or an eigenvector.

In some embodiments, the real value corresponding to the output information includes a channel matrix and/or an eigenvector.

In some embodiments, the first information includes at least one of the following: antenna configuration information of the terminal device, moving speed information of the terminal device, or expected time information of the terminal device.

In some embodiments, the antenna configuration information includes at least one of the following: the number of horizontal panels and the number of vertical panels; the horizontal panel spacing and the vertical panel spacing; the number of horizontal antennas within the panel and the number of vertical antennas within the panel; the horizontal antenna spacing within the panel and the vertical antenna spacing within the panel; the mapping relationship between the antenna and the TxRU; or the antenna polarization direction.

In some embodiments, the time information expected by the terminal device includes at least one of the following: the first time information for performing CSI-RS measurement, the second time information for performing CSI prediction, or the interval between the first time for performing CSI-RS measurement and the second time for performing CSI prediction.

In some embodiments, the CSI resource configuration information is used to configure one or more CSI-RS resources.

In some embodiments, the CSI-RS resources include at least one of the following: periodic CSI-RS resources, semi-static CSI-RS resources, or aperiodic CSI-RS resources with equal intervals in the time domain.

In some embodiments, the CSI-RS resources are used for reporting the training data and reporting channel state information.

In some embodiments, the CSI-RS resources are used for reporting the training data and monitoring the performance of the model.

In some embodiments, the performance monitoring result of the model is lower than a first threshold, and the training data is reported; and/or the performance monitoring result of the model is higher than a second threshold, and the training data is reported.

In some embodiments, the CSI-RS resources are configured according to at least one of the following: a scenario of the network device, a moving speed of the terminal device, or a desired configuration of the terminal device.

In some embodiments, the number of moments and/or intervals of the CSI-RS resources are configured according to at least one of the following: a scenario of the network device, a moving speed of the terminal device, or a desired configuration of the terminal device.

In some embodiments, the CSI-RS resources include a first resource corresponding to a first CSI-RS and a second resource corresponding to a second CSI-RS, the measurement result of the first CSI-RS is used to generate input information of the model, and the measurement result of the second CSI-RS is used to generate a true value corresponding to the output information of the model.

In some embodiments, the first resource and the second resource are configured separately.

In some embodiments, the first resource is associated with the second resource.

In some embodiments, the first resource and the second resource are jointly configured.

In some embodiments, the measurement result of the first CSI-RS and the measurement result of the second CSI-RS correspond to the same receive beam.

In some embodiments, the CSI reporting configuration information and/or the CSI resource configuration information includes first indication information for instructing reporting of the training data.

In some embodiments, the training data is reported via non-layer 1 signaling, and the first indication information is carried on radio resource control (RRC) signaling.

In some embodiments, the training data is reported via layer 1 signaling, and the first indication information is carried on at least one of the following: radio resource control (RRC) signaling, media access control layer control element (MAC CE), or downlink control information (DCI).

In some embodiments, the training data is reported once or divided into multiple reports.

In some embodiments, the network device further sends second information of the network device to the terminal device, so that the terminal device classifies and/or labels the training data according to the second information and/or the first information of the terminal device.

In some embodiments, the training data includes first information of the terminal device; the network device classifies and/or labels the training data according to second information on the network side and/or the first information.

In some embodiments, the second information on the network side includes at least one of the following: Configuration information, scenario information of network equipment, reference signal period, cell/site identifier, carrier frequency, frequency domain granularity, or subcarrier spacing.

In some embodiments, the antenna configuration information includes at least one of the following: the number of horizontal panels and the number of vertical panels; the horizontal panel spacing and the vertical panel spacing; the number of horizontal antennas within the panel and the number of vertical antennas within the panel; the horizontal antenna spacing within the panel and the vertical antenna spacing within the panel; the mapping relationship between the antenna and the TxRU; or the antenna polarization direction.

In some embodiments, the scene information includes: indoor or outdoor, and/or line-of-sight or non-line-of-sight.

In some embodiments, the network device can use the classified training data to train the AI/ML model and send relevant information of the trained AI/ML model to the terminal device.

In some embodiments, the network device may also indicate to the terminal device classification information corresponding to the relevant information of the AI/ML model, for example, indicating a classification label and/or first information and/or second information corresponding to the relevant information of the AI/ML model.

The above embodiments are merely exemplary of the present invention, but the present invention is not limited thereto. Appropriate modifications may be made based on the above embodiments. For example, the above embodiments may be used alone, or one or more of the above embodiments may be combined.

According to the above embodiment, the network device sends CSI resource configuration information and/or CSI reporting configuration information to the terminal device and receives training data for the AI/ML model reported by the terminal device. This allows the collection of training data for the AI/ML model, helping to improve the accuracy and performance of the AI/ML model.

Embodiments of the third aspect

The embodiment of the present application provides an information processing device. The device may be, for example, a terminal device, or one or more components or assemblies configured in the terminal device, which corresponds to the embodiment of the first aspect, and the same contents as the embodiment of the first aspect are not repeated here.

FIG4 is a schematic diagram of an information processing device according to an embodiment of the present application. As shown in FIG4 , the information processing device 400 includes:

A receiving unit 401, which receives CSI resource configuration information and/or CSI reporting configuration information from a network device; and

A sending unit 402 reports training data for an AI/ML model according to the CSI resource configuration information and/or the CSI reporting configuration information.

In some embodiments, the training data includes at least one of the following: input information of the model; a true value corresponding to output information of the model; or first information of the terminal device.

In some embodiments, the input information includes a channel matrix and/or an eigenvector; and/or, the true value corresponding to the output information includes a channel matrix and/or an eigenvector; and/or, the first information includes at least one of the following: antenna configuration information of the terminal device, moving speed information of the terminal device, or expected time information of the terminal device.

In some embodiments, the antenna configuration information includes at least one of the following: the number of horizontal panels and the number of vertical panels; the horizontal panel spacing and the vertical panel spacing; the number of horizontal antennas within a panel and the number of vertical antennas within a panel; the horizontal antenna spacing within a panel and the vertical antenna spacing within a panel; a mapping relationship between antennas and TxRUs; or antenna polarization direction;

and/or

The time information expected by the terminal device includes at least one of the following: the first time information for performing CSI-RS measurement, the second time information for performing CSI prediction, or the interval between the first time for performing CSI-RS measurement and the second time for performing CSI prediction.

In some embodiments, the CSI resource configuration information is used to configure one or more CSI-RS resources.

In some embodiments, the CSI-RS resources include at least one of the following: periodic CSI-RS resources, semi-static CSI-RS resources, or aperiodic CSI-RS resources with equal intervals in the time domain.

In some embodiments, the CSI-RS resources are used for reporting the training data and reporting channel state information.

In some embodiments, the CSI-RS resources are used for reporting the training data and monitoring the performance of the model.

In some embodiments, the performance monitoring result of the model is lower than a first threshold, and the training data is reported; and/or the performance monitoring result of the model is higher than a second threshold, and the training data is reported.

In some embodiments, the CSI-RS resources are configured according to at least one of the following: a scenario of the network device, a moving speed of the terminal device, or a desired configuration of the terminal device.

In some embodiments, the number of moments and/or intervals of the CSI-RS resources are configured according to at least one of the following: a scenario of the network device, a moving speed of the terminal device, or a desired configuration of the terminal device.

In some embodiments, the CSI-RS resources include a first resource corresponding to a first CSI-RS and a second resource corresponding to a second CSI-RS, and a measurement result of the first CSI-RS is used to generate input information of the model. The measurement result of the second CSI-RS is used to generate a true value corresponding to the output information of the model.

In some embodiments, the first resource and the second resource are configured separately; and/or, the first resource is associated with the second resource; and/or, the first resource and the second resource are jointly configured.

In some embodiments, the measurement result of the first CSI-RS and the measurement result of the second CSI-RS correspond to the same receive beam.

In some embodiments, the CSI reporting configuration information and/or the CSI resource configuration information includes first indication information for instructing reporting of the training data.

In some embodiments, the training data is reported via non-layer 1 signaling, and the first indication information is carried on radio resource control (RRC) signaling; and/or, the training data is reported via layer 1 signaling, and the first indication information is carried on at least one of the following: radio resource control (RRC) signaling, media access control layer control element (MAC CE), or downlink control information (DCI).

In some embodiments, the training data is reported once or divided into multiple reports.

In some embodiments, the receiving unit further receives second information from the network device; the apparatus 400 further includes:

The processing unit 403 classifies and/or labels the training data according to the second information and/or the first information of the terminal device, and the sending unit reports the classified and/or labeled training data.

The above embodiments are merely exemplary of the present invention, but the present invention is not limited thereto. Appropriate modifications may be made based on the above embodiments. For example, the above embodiments may be used alone, or one or more of the above embodiments may be combined.

According to the above embodiment, the terminal device receives CSI resource configuration information and/or CSI reporting configuration information from the network device and reports training data for the AI/ML model based on the CSI resource configuration information and/or CSI reporting configuration information. This enables the collection of training data for the AI/ML model, helping to improve the accuracy and performance of the AI/ML model.

Embodiments of the fourth aspect

The embodiment of the present application provides an information processing device. The device may be, for example, a network device, or one or more components or assemblies configured in the network device, which corresponds to the embodiment of the second aspect, and the same contents as the embodiment of the second aspect are not repeated here.

FIG5 is another schematic diagram of the information processing device according to an embodiment of the present application. As shown in FIG5 , the information processing device 500 includes:

A sending unit 501, which sends CSI resource configuration information and/or CSI reporting configuration information to a terminal device; and

A receiving unit 502 receives training data for the AI/ML model reported by the terminal device.

In some embodiments, the training data includes at least one of the following: input information of the model; a true value corresponding to output information of the model; or first information of the terminal device.

In some embodiments, the input information includes a channel matrix and/or an eigenvector.

In some embodiments, the real value corresponding to the output information includes a channel matrix and/or an eigenvector.

In some embodiments, the first information includes at least one of the following: antenna configuration information of the terminal device, moving speed information of the terminal device, or expected time information of the terminal device.

In some embodiments, the antenna configuration information includes at least one of the following: the number of horizontal panels and the number of vertical panels; the horizontal panel spacing and the vertical panel spacing; the number of horizontal antennas within the panel and the number of vertical antennas within the panel; the horizontal antenna spacing within the panel and the vertical antenna spacing within the panel; the mapping relationship between the antenna and the TxRU; or the antenna polarization direction.

In some embodiments, the time information expected by the terminal device includes at least one of the following: the first time information for performing CSI-RS measurement, the second time information for performing CSI prediction, or the interval between the first time for performing CSI-RS measurement and the second time for performing CSI prediction.

In some embodiments, the CSI resource configuration information is used to configure one or more CSI-RS resources.

In some embodiments, the CSI-RS resources include at least one of the following: periodic CSI-RS resources, semi-static CSI-RS resources, or aperiodic CSI-RS resources with equal intervals in the time domain.

In some embodiments, the CSI-RS resources are used for reporting the training data and reporting channel state information.

In some embodiments, the CSI-RS resources are used for reporting the training data and monitoring the performance of the model.

In some embodiments, the performance monitoring result of the model is lower than a first threshold, and the training data is reported; and/or the performance monitoring result of the model is higher than a second threshold, and the training data is reported.

In some embodiments, the CSI-RS resources are configured according to at least one of the following: a scenario of the network device, a moving speed of the terminal device, or a desired configuration of the terminal device.

In some embodiments, the number and/or interval of the CSI-RS resources are based on at least one of the following: Row configuration: the scenario of the network device, the moving speed of the terminal device, or the desired configuration of the terminal device.

In some embodiments, the CSI-RS resources include a first resource corresponding to a first CSI-RS and a second resource corresponding to a second CSI-RS, the measurement result of the first CSI-RS is used to generate input information of the model, and the measurement result of the second CSI-RS is used to generate a true value corresponding to the output information of the model.

In some embodiments, the first resource and the second resource are configured separately.

In some embodiments, the first resource is associated with the second resource.

In some embodiments, the first resource and the second resource are jointly configured.

In some embodiments, the measurement result of the first CSI-RS and the measurement result of the second CSI-RS correspond to the same receive beam.

In some embodiments, the CSI reporting configuration information and/or the CSI resource configuration information includes first indication information for instructing reporting of the training data.

In some embodiments, the training data is reported via non-layer 1 signaling, and the first indication information is carried on radio resource control (RRC) signaling.

In some embodiments, the training data is reported via layer 1 signaling, and the first indication information is carried on at least one of the following: radio resource control (RRC) signaling, media access control layer control element (MAC CE), or downlink control information (DCI).

In some embodiments, the training data is reported once or divided into multiple reports.

In some embodiments, the sending unit 501 further sends the second information of the network device to the terminal device, so that the terminal device can classify and/or mark the training data according to the second information and/or the first information of the terminal device.

In some embodiments, the training data includes first information of the terminal device; the apparatus 500 further includes: a processing unit 503, which classifies and/or labels the training data according to second information on the network side and/or the first information.

In some embodiments, the second information on the network side includes at least one of the following: antenna configuration information of the network device, scenario information of the network device, reference signal period, cell/site identification, carrier frequency, frequency domain granularity, or subcarrier spacing.

In some embodiments, the antenna configuration information includes at least one of the following: the number of horizontal panels and the number of vertical panels; the horizontal panel spacing and the vertical panel spacing; the number of horizontal antennas within the panel and the number of vertical antennas within the panel; the horizontal antenna spacing within the panel and the vertical antenna spacing within the panel; the mapping relationship between the antenna and the TxRU; or the antenna polarization direction.

In some embodiments, the scene information includes: indoor or outdoor, and/or line-of-sight or non-line-of-sight.

The above embodiments are merely exemplary of the present invention, but the present invention is not limited thereto. Appropriate modifications may be made based on the above embodiments. For example, the above embodiments may be used alone, or one or more of the above embodiments may be combined.

According to the above embodiment, the network device sends CSI resource configuration information and/or CSI reporting configuration information to the terminal device and receives training data for the AI/ML model reported by the terminal device. This allows the collection of training data for the AI/ML model, helping to improve the accuracy and performance of the AI/ML model.

Embodiments of the fifth aspect

An embodiment of the present application also provides a communication system, and reference may be made to FIG1 . The contents that are the same as those in the embodiments of the first to fourth aspects will not be repeated.

In some embodiments, the communication system 100 may include at least: a network device and a terminal device. The network device sends CSI resource configuration information and/or CSI reporting configuration information to the terminal device and receives training data for an AI/ML model reported by the terminal device; the terminal device receives the CSI resource configuration information and/or CSI reporting configuration information and reports the training data based on the CSI resource configuration information and/or CSI reporting configuration information.

An embodiment of the present application further provides a network device, which may be, for example, a base station, but the present application is not limited thereto and may also be other network devices.

Figure 6 is a schematic diagram illustrating the structure of a network device according to an embodiment of the present application. As shown in Figure 6 , network device 600 may include a processor 610 (e.g., a central processing unit (CPU)) and a memory 620 ; the memory 620 is coupled to the processor 610 . The memory 620 may store various data and may also store an information processing program 630 , which is executed under the control of the processor 610 .

For example, the processor 610 may be configured to execute a program to implement the operation of the network device in the method according to the embodiment of the second aspect. For example, the processor 610 may be configured to perform the following control: the network device sends CSI resource configuration information and/or CSI reporting configuration information to the terminal device; and the network device receives training data for the AI/ML model reported by the terminal device.

In addition, as shown in FIG6 , the network device 600 may further include: a transceiver (receiver and/or transmitter) 640 and an antenna 650, etc.; wherein, the functions of the above components are similar to those in the related art and are not described here in detail. It is worth noting that the network device 600 does not necessarily have to include all the components shown in FIG6 ; in addition, the network device 600 may also include components not shown in FIG6 , and reference may be made to related art for details.

The embodiment of the present application also provides a terminal device, but the present application is not limited thereto and may also be other devices.

Figure 7 is a schematic diagram of a terminal device according to an embodiment of the present application. As shown in Figure 7 , terminal device 700 may include a processor 710 and a memory 720. Memory 720 stores data and programs and is coupled to processor 710. It should be noted that this diagram is exemplary; other types of structures may be used to supplement or replace this structure to implement telecommunication or other functions.

For example, the processor 710 may be configured to execute a program to implement the method according to the embodiment of the first aspect. For example, the processor 710 may be configured to perform the following control: the terminal device receives CSI resource configuration information and/or CSI reporting configuration information from the network device; and the terminal device reports training data for the AI/ML model based on the CSI resource configuration information and/or the CSI reporting configuration information.

As shown in Figure 7 , the terminal device 700 may further include: a communication module 730, an input unit 740, a display 750, and a power supply 760. The functions of these components are similar to those in the related art and are not described in detail here. It is worth noting that the terminal device 700 does not necessarily include all of the components shown in Figure 7 , and the above components are not essential. Furthermore, the terminal device 700 may also include components not shown in Figure 7 , for which reference may be made to the related art.

An embodiment of the present application further provides a computer program, wherein when the program is executed in a terminal device, the program causes the terminal device to execute the method described in the embodiment of the first aspect.

An embodiment of the present application further provides a storage medium storing a computer program, wherein the computer program enables a terminal device to execute the method described in the embodiment of the first aspect.

An embodiment of the present application further provides a computer program, wherein when the program is executed in a network device, the program causes the network device to execute the method described in the embodiment of the second aspect.

An embodiment of the present application further provides a storage medium storing a computer program, wherein the computer program enables a network device to execute the method described in the embodiment of the second aspect.

The above devices and methods of the present application can be implemented by hardware or by a combination of hardware and software. The present application relates to such a computer-readable program that, when executed by a logic component, enables the logic component to implement the devices or components described above, or enables the logic component to implement the various methods or steps described above. The present application also relates to a storage medium for storing the above program, such as a hard disk, a magnetic disk, an optical disk, a DVD, a flash memory, etc.

The method/device described in conjunction with the embodiments of the present application may be directly embodied as hardware, a software module executed by a processor, or a combination of the two. For example, one or more of the functional block diagrams shown in the figure and/or one or more of the functional block diagrams The multiple combinations can correspond to either software modules or hardware modules of a computer program flow. These software modules can correspond to the steps shown in the figure. These hardware modules can be implemented by solidifying these software modules using, for example, a field programmable gate array (FPGA).

The software module may be located in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art. A storage medium may be coupled to a processor so that the processor can read information from the storage medium and write information to the storage medium; or the storage medium may be an integral part of the processor. The processor and the storage medium may be located in an ASIC. The software module may be stored in the memory of the mobile terminal or in a memory card that can be inserted into the mobile terminal. For example, if the device (such as a mobile terminal) uses a large-capacity MEGA-SIM card or a large-capacity flash memory device, the software module may be stored in the MEGA-SIM card or the large-capacity flash memory device.

One or more of the functional blocks and/or one or more combinations of functional blocks described in the accompanying drawings may be implemented as a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic device, a discrete gate or transistor logic device, a discrete hardware component, or any appropriate combination thereof for performing the functions described in this application. One or more of the functional blocks and/or one or more combinations of functional blocks described in the accompanying drawings may also be implemented as a combination of computing devices, such as a combination of a DSP and a microprocessor, multiple microprocessors, one or more microprocessors in communication with a DSP, or any other such configuration.

The present application has been described above in conjunction with specific embodiments. However, those skilled in the art should understand that these descriptions are merely illustrative and are not intended to limit the scope of protection of the present application. Those skilled in the art may make various modifications and variations to the present application based on the spirit and principles of the present application, and such modifications and variations are also within the scope of the present application.

Regarding the implementation methods including the above embodiments, the following additional notes are also disclosed:

1. An information processing device, configured in a terminal device, wherein the device comprises:

a receiving unit configured to receive CSI resource configuration information and/or CSI reporting configuration information from a network device; and

A sending unit, which reports training data for an AI/ML model based on the CSI resource configuration information and/or the CSI reporting configuration information.

2. The apparatus according to Note 1, wherein the number of times and/or intervals of the CSI-RS resources are based on Configure at least one of the following:

The scenario of the network device, the moving speed of the terminal device, or the expected configuration of the terminal device.

3. An information processing device, configured in a network device, wherein the device comprises:

a sending unit, configured to send CSI resource configuration information and/or CSI reporting configuration information to a terminal device; and

A receiving unit receives training data for the AI/ML model reported by the terminal device.

4. The apparatus according to Supplement 3, wherein the second information on the network side includes at least one of the following:

Antenna configuration information of the network device, scenario information of the network device, reference signal period, cell/site identifier, carrier frequency, frequency domain granularity, or subcarrier spacing.

5. The device according to Supplement 4, wherein:

The antenna configuration information includes at least one of the following:

The number of horizontal panels and the number of vertical panels;

Horizontal panel spacing and vertical panel spacing;

The number of horizontal antennas in the panel and the number of vertical antennas in the panel;

Horizontal antenna spacing within the panel and vertical antenna spacing within the panel;

The mapping relationship between antennas and TxRUs; or

Antenna polarization direction.

6. The device according to Supplement 4, wherein:

The scene information includes: indoor or outdoor, and/or line-of-sight or non-line-of-sight.

Claims (20)

An information processing device, configured in a terminal device, wherein the device comprises: a receiving unit configured to receive channel state information (CSI) resource configuration information and/or CSI reporting configuration information from a network device; and A sending unit, which reports training data for an AI/ML model based on the CSI resource configuration information and/or the CSI reporting configuration information. The apparatus of claim 1, wherein the training data comprises at least one of the following: input information of the model; The true value corresponding to the output information of the model; or The first information of the terminal device. The device according to claim 2, wherein The input information includes a channel matrix and/or an eigenvector; and/or The real value corresponding to the output information includes a channel matrix and/or a eigenvector; and/or The first information includes at least one of the following: antenna configuration information of the terminal device, moving speed information of the terminal device, or expected time information of the terminal device. The device according to claim 3, wherein The antenna configuration information includes at least one of the following: the number of horizontal panels and the number of vertical panels; the horizontal panel spacing and the vertical panel spacing; the number of horizontal antennas within a panel and the number of vertical antennas within a panel; the horizontal antenna spacing within a panel and the vertical antenna spacing within a panel; the mapping relationship between antennas and TxRUs; or the antenna polarization direction; and/or The time information expected by the terminal device includes at least one of the following: the first time information for performing CSI-RS measurement, the second time information for performing CSI prediction, or the interval between the first time for performing CSI-RS measurement and the second time for performing CSI prediction. The device according to claim 1, wherein The CSI resource configuration information is used to configure one or more CSI-RS resources. The device according to claim 5, wherein The CSI-RS resources include at least one of the following: periodic CSI-RS resources, semi-static CSI-RS resources, or aperiodic CSI-RS resources with equal intervals in the time domain. The device according to claim 5, wherein The CSI-RS resources are used for reporting the training data and reporting channel state information. The device according to claim 5, wherein The CSI-RS resources are used for reporting the training data and monitoring the performance of the model. The device according to claim 8, wherein The performance monitoring result of the model is lower than a first threshold, and the training data is reported; and/or The performance monitoring result of the model is higher than a second threshold, and the training data is reported. The apparatus according to claim 5, wherein the CSI-RS resource is configured according to at least one of the following: The scenario of the network device, the moving speed of the terminal device, or the expected configuration of the terminal device. The device according to claim 5, wherein The CSI-RS resources include a first resource corresponding to a first CSI-RS and a second resource corresponding to a second CSI-RS. The measurement result of the first CSI-RS is used to generate input information of the model, and the measurement result of the second CSI-RS is used to generate a true value corresponding to the output information of the model. The device according to claim 11, wherein The first resource and the second resource are configured separately; and/or The first resource is associated with the second resource; and/or The first resource and the second resource are jointly configured. The device according to claim 11, wherein The measurement result of the first CSI-RS and the measurement result of the second CSI-RS correspond to the same receive beam. The device according to claim 1, wherein The CSI reporting configuration information and/or the CSI resource configuration information includes first indication information for instructing reporting of the training data. The device according to claim 14, wherein The training data is reported via non-layer 1 signaling, and the first indication information is carried on radio resource control (RRC) signaling; and/or The training data is reported via layer 1 signaling, and the first indication information is carried on at least one of the following: radio resource control (RRC) signaling, media access control layer control element (MAC CE), or downlink control element (DCE). Direct Control Information (DCI). The device according to claim 15, wherein The training data is reported once or divided into multiple reports. The device according to claim 1, wherein The receiving unit further receives second information from the network device; The device further comprises: A processing unit classifies and/or labels the training data according to the second information and/or the first information of the terminal device, and the sending unit reports the classified and/or labeled training data. An information processing device, configured in a network device, wherein the device comprises: a sending unit, configured to send CSI resource configuration information and/or CSI reporting configuration information to a terminal device; and A receiving unit receives training data for the AI/ML model reported by the terminal device. The device according to claim 18, wherein The training data includes first information of the terminal device; The device further comprises: A processing unit is configured to classify and/or label the training data according to the second information on the network side and/or the first information. A communication system, wherein the system includes a network device and a terminal device, The network device sends CSI resource configuration information and/or CSI reporting configuration information to the terminal device, and receives training data for the AI/ML model reported by the terminal device; The terminal device receives the CSI resource configuration information and/or CSI reporting configuration information, and reports the training data according to the CSI resource configuration information and/or CSI reporting configuration information.