WO2025065195A1 - Processing method, and device and storage medium - Google Patents

Abstract

The present disclosure relates to a processing method, and a terminal, a network device and a storage medium. The method comprises: acquiring first input data and first output data, wherein the first input data comprises a compressed measurement result; training a first decoding model on the basis of the first input data and the first output data, so as to obtain a trained first decoding model; and on the basis of the trained first decoding model, acquiring a similarity difference value, wherein the similarity difference value is used for indicating the difference between the first decoding model and a second decoding model. In the embodiments, a method for acquiring the difference between decoding models of two devices is provided, such that the operation performance of the decoding model of a second device can be supervised by means of a first device, and the operation performance of the decoding model of the second device can be determined without data transmission, thereby reducing the data transmission volume, ensuring a reduction in the complexity in monitoring the operation of the model of the second device, ensuring the normal operation of the model, and thus ensuring the reliability of communication.

Description

Processing method, device and storage medium Technical Field

The present disclosure relates to the field of communication technology, and in particular to a processing method, a device, and a storage medium.

Background Art

With the rapid development of mobile communication technology, communication between devices can be carried out through data transmission through model encoding to achieve the purpose of data compression, thereby saving transmission resources. This puts forward the need to supervise whether the operating performance of the model meets the requirements and realize the supervision of the model.

Summary of the invention

The present disclosure solves the problem of supervising the operating performance of the model of the second device through the first device, thereby reducing the complexity of monitoring the operating conditions of the model of the second device, ensuring the normal operation of the model, and further ensuring the reliability of communication.

The embodiments of the present disclosure provide a processing method, a device, and a storage medium.

According to a first aspect of an embodiment of the present disclosure, a processing method is proposed, the method comprising:

Acquire first input data and first output data, wherein the first input data includes a compressed measurement result; the first output data includes an original measurement result or a second measurement result restored after a second decoding model decodes the first measurement result of the first device, and the second decoding model is set on the second device;

Training a first decoding model based on the first input data and the first output data to obtain a trained first decoding model, wherein the first decoding model is set in the first device;

Based on the trained first decoding model, a similarity difference value is obtained, where the similarity difference value is used to indicate the difference between the first decoding model and the second decoding model.

According to a second aspect of an embodiment of the present disclosure, a processing method is proposed, the method comprising:

The second device sends first input data and first output data, wherein the first input data includes a compressed measurement result; the first output data includes an original measurement result or a second measurement result restored after a second decoding model decodes the first measurement result of the first device, and the second decoding model is set on the second device;

The first input data and the first output data are used to train a first decoding model, and the first decoding model is set in the first device;

The trained first decoding model is used to obtain a similarity difference value, and the similarity difference value is used to indicate the difference between the first decoding model and the second decoding model.

According to a third aspect of the embodiments of the present disclosure, a processing method is proposed, the method comprising:

The second device sends the first input data and the first output data;

The first device acquires first input data and first output data, wherein the first input data includes a compressed measurement result; the first output data includes an original measurement result or a second measurement result restored after a second decoding model decodes the first measurement result of the first device, and the second decoding model is set on the second device;

The first device trains a first decoding model based on the first input data and the first output data to obtain a training a first decoding model after the first decoding model is set in the first device;

The first device obtains a similarity difference value based on the trained first decoding model, where the similarity difference value is used to indicate a difference between the first decoding model and the second decoding model.

According to a fourth aspect of an embodiment of the present disclosure, a first device is provided, including:

a processing module, configured to obtain first input data and first output data, wherein the first input data includes a compressed measurement result; the first output data includes an original measurement result or a second measurement result restored after a second decoding model decodes the first measurement result of the first device, and the second decoding model is set on the second device;

The processing module is further used to train a first decoding model based on the first input data and the first output data to obtain a trained first decoding model, wherein the first decoding model is set in the first device;

The processing module is further used to obtain a similarity difference value based on the trained first decoding model, where the similarity difference value is used to indicate the difference between the first decoding model and the second decoding model.

According to a fifth aspect of an embodiment of the present disclosure, a second device is provided, including:

a transceiver module, configured to send first input data and first output data, wherein the first input data includes a compressed measurement result; and the first output data includes an original measurement result or a second measurement result restored after a second decoding model decodes the first measurement result of the first device, wherein the second decoding model is set on the second device;

The first input data and the first output data are used to train a first decoding model, and the first decoding model is set in the first device;

The trained first decoding model is used to obtain a similarity difference value, and the similarity difference value is used to indicate the difference between the first decoding model and the second decoding model.

According to a sixth aspect of an embodiment of the present disclosure, a first device is provided, including:

one or more processors;

The first device is used to execute any method described in the first aspect.

According to a seventh aspect of an embodiment of the present disclosure, a second device is provided, including:

one or more processors;

The second device is used to execute any method described in the second aspect.

According to an eighth aspect of an embodiment of the present disclosure, a communication system is provided, including:

A first device and a second device, wherein the first device is configured to implement the processing method described in the first aspect, and the second device is configured to implement the processing method described in the second aspect.

According to a ninth aspect of an embodiment of the present disclosure, a storage medium is proposed, wherein the storage medium stores instructions, and when the instructions are executed on a communication device, the communication device executes a method as described in any one of the first aspect or the second aspect.

BRIEF DESCRIPTION OF THE DRAWINGS

The drawings described herein are used to provide a further understanding of the embodiments of the present disclosure and constitute a part of the present disclosure. The illustrative embodiments of the embodiments of the present disclosure and their descriptions are used to explain the embodiments of the present disclosure and do not constitute an improper limitation on the embodiments of the present disclosure. In the drawings:

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

FIG2 is an interactive schematic diagram of a processing method according to an embodiment of the present disclosure;

FIG3A is a schematic flow chart of a processing method according to an embodiment of the present disclosure;

FIG3B is a flow chart of a processing method according to an embodiment of the present disclosure;

FIG4 is a schematic flow chart of a processing method according to an embodiment of the present disclosure;

FIG5 is a schematic flow chart of a processing method according to an embodiment of the present disclosure;

FIG6A is a schematic diagram of the structure of a terminal provided in an embodiment of the present disclosure;

FIG6B is a schematic diagram of the structure of a network device proposed in an embodiment of the present disclosure;

FIG7A is a schematic diagram of the structure of a communication device provided in an embodiment of the present disclosure;

FIG. 7B is a schematic diagram of the structure of a chip proposed in an embodiment of the present disclosure.

DETAILED DESCRIPTION

The present disclosure provides a processing method, a terminal and a storage medium.

In a first aspect, an embodiment of the present disclosure provides a processing method, the method comprising:

Acquire first input data and first output data, wherein the first input data includes a compressed measurement result; the first output data includes an original measurement result or a second measurement result restored after a second decoding model decodes the first measurement result of the first device, and the second decoding model is set on the second device;

Training a first decoding model based on the first input data and the first output data to obtain a trained first decoding model, wherein the first decoding model is set in the first device;

Based on the trained first decoding model, a similarity difference value is obtained, where the similarity difference value is used to indicate the difference between the first decoding model and the second decoding model.

In some embodiments, the original measurement result can be understood as the real CSI (Channel State Information, channel state reference signal), that is, Ground-truth CSI. When the ground-truth CSI is transmitted, its representation method can be Float 32 or high-resolution codebook form.

In the above embodiment, a method for obtaining the difference in decoding models between two devices is provided, and the operating performance of the decoding model of the second device can be supervised by the first device. The operating performance of the decoding model of the second device can be determined without data transmission, thereby reducing the amount of data transmission, reducing the complexity of monitoring the operating status of the model of the second device, ensuring the normal operation of the model, and thus ensuring the reliability of communication.

In the above embodiment, the first decoding model of the first device can be trained by obtaining compressed measurement results and original measurement results, so as to determine the similarity difference according to the trained decoding model, and then the operating performance of the decoding model of the second device can be monitored according to the similarity difference to ensure the accuracy of monitoring.

In combination with some embodiments of the first aspect, in some embodiments, the similarity is obtained based on the trained first decoding model. Degree difference, including:

Determine at least one first training similarity based on the trained first decoding model, where the first training similarity is used to indicate the similarity between a third measurement result obtained by decoding the first measurement result by the trained first decoding model and the original measurement result;

Determine at least one second training similarity based on the second decoding model, where the second training similarity is used to indicate a similarity between a fourth measurement result obtained by decoding the first measurement result by the second decoding model and an original measurement result;

The similarity difference is determined based on the at least one first training similarity and the at least one second training similarity.

In the above embodiment, the similarity of decoding by the first decoding model and the similarity of the second decoding model can be obtained respectively by running the first decoding model and the second decoding model, and then the similarity difference can be determined according to the similarities of the two decoding models to ensure the accuracy of the obtained similarity.

In combination with some embodiments of the first aspect, the determining the similarity difference based on the at least one first training similarity and the at least one second training similarity includes:

Obtaining at least one difference between at least one of the first training similarities and a corresponding second training similarity;

An average value of the at least one difference value is determined as the similarity difference value.

In the above embodiment, the similarity difference is determined by obtaining the average value of the difference, thereby ensuring the accuracy of the obtained similarity difference.

In combination with some embodiments of the first aspect, in some embodiments, the variance or standard deviation of the difference is less than a difference threshold.

In the above embodiment, the variance or standard deviation of the obtained difference is smaller than the difference threshold, which ensures the accuracy and stability of the obtained data, and further ensures the accuracy of the obtained similarity difference.

In combination with some embodiments of the first aspect, determining at least one first training similarity based on the trained first decoding model includes:

Encoding the fifth measurement result of the first device by using a first coding model to obtain the first measurement result;

Decoding the first measurement result using the trained first decoding model to obtain a third measurement result;

The first training similarity is determined based on the fifth measurement result and the third measurement result.

In the above embodiment, the first training similarity can be obtained by encoding and decoding the measurement result, thereby ensuring the accuracy of the obtained first training similarity.

In combination with some embodiments of the first aspect, in some embodiments, determining at least one second training similarity based on the second decoding model includes:

Encoding the fifth measurement result of the first device by using a first coding model to obtain the first measurement result;

Decoding the first measurement result using the second decoding model to obtain the fourth measurement result;

The second training similarity is determined based on the fifth measurement result and the fourth measurement result.

In the above embodiment, the second training similarity can be obtained by encoding and decoding the measurement result, thereby ensuring the accuracy of the obtained second training similarity.

In conjunction with some embodiments of the first aspect, in some embodiments, obtaining the first input data and the first output data includes:

First information sent by a second device is received, where the first information includes the first input data and the first output data.

In conjunction with some embodiments of the first aspect, in some embodiments, the first information further includes at least one of the following:

Dataset identifier;

whether the first input data is quantitative data;

a quantization method for quantizing the first input data;

Scene information, where the scene information is used to indicate a scene in which the first information is located;

Configuration information, where the configuration information is used to indicate a configuration for the first device to perform measurement.

In a second aspect, an embodiment of the present disclosure provides a processing method, the method comprising:

Sending first input data and first output data, wherein the first input data includes a compressed measurement result; the first output data includes an original measurement result or a second measurement result restored after a second decoding model decodes the first measurement result of the first device, and the second decoding model is set on the second device;

The first input data and the first output data are used to train a first decoding model, and the first decoding model is set in the first device;

The trained first decoding model is used to obtain a similarity difference value, and the similarity difference value is used to indicate the difference between the first decoding model and the second decoding model.

In combination with some embodiments of the first aspect, in some embodiments, the similarity difference is obtained based on the trained first decoding model;

The trained first decoding model is obtained by training the first decoding model based on the first input data and the first output data;

The first input data includes a compressed measurement result; the first output data includes an original measurement result or a second measurement result restored after the second decoding model decodes the first measurement result of the first device.

In combination with some embodiments of the first aspect, in some embodiments, obtaining the similarity difference based on the trained first decoding model includes:

The similarity difference is determined based on the at least one first training similarity and the at least one second training similarity;

The at least one second training similarity is determined based on the second decoding model, and the second training similarity is used to indicate the similarity between the fourth measurement result obtained by decoding the first measurement result by the second decoding model and the original measurement result; the at least one first training similarity is determined based on the trained first decoding model, and the first training similarity is used to indicate the similarity between the third measurement result obtained by decoding the first measurement result by the trained first decoding model and the original measurement result.

In conjunction with some embodiments of the first aspect, in some embodiments, the similarity difference is an average value of the at least one difference;

The at least one difference value is a difference value between at least one of the first training similarities and a corresponding second training similarity.

In combination with some embodiments of the first aspect, in some embodiments, the variance or standard deviation of the difference is less than a difference threshold.

In combination with some embodiments of the first aspect, in some embodiments, the first training similarity is determined based on the fifth measurement result and the third measurement result; the first measurement result is obtained by encoding the fifth measurement result of the first device using a first coding model;

The third measurement result is obtained by decoding the first measurement result using the trained first decoding model.

In combination with some embodiments of the first aspect, in some embodiments, the second training similarity is determined based on the fifth measurement result and the fourth measurement result;

The first measurement result is obtained by encoding a fifth measurement result of the first device using a first coding model;

The fourth measurement result is obtained by decoding the first measurement result using the second decoding model.

In combination with some embodiments of the first aspect, in some embodiments, a second device sends the first information to the first device, and the first information includes the first input data and the first output data.

In conjunction with some embodiments of the first aspect, in some embodiments, the first information further includes at least one of the following:

Dataset identifier;

whether the first input data is quantitative data;

a quantization method for quantizing the first input data;

Scene information, where the scene information is used to indicate a scene in which the first information is located;

Configuration information, where the configuration information is used to indicate a configuration for the first device to perform measurement.

In a third aspect, an embodiment of the present disclosure provides a processing method, the method comprising:

The first device determines whether the operating performance of the second decoding model of the second device is higher than the second similarity based on the first similarity and the similarity difference of the first decoding model of the first device, the first decoding model and the second decoding model being used to decode the first measurement result of the first device;

Whether the operating performance of the second decoding model of the second device is higher than the second similarity is determined by the first similarity and the similarity difference of the first decoding model of the first device, and the first decoding model and the second decoding model are used to decode the first measurement result of the first device.

The embodiments of the present disclosure provide processing methods, terminals, network devices, and storage media. In some embodiments, the terms processing method, information processing method, indication method, etc. can be interchangeable, the terms communication device, information processing device, indication device, etc. can be interchangeable, and the terms information processing system, communication system, etc. can be interchangeable.

The embodiments of the present disclosure are not exhaustive, but are only illustrative of some embodiments, and are not intended to be a specific limitation on the scope of protection of the present disclosure. In the absence of contradiction, each step in a certain embodiment can be implemented as an independent embodiment, and the steps can be arbitrarily combined. For example, a solution after removing some steps in a certain embodiment can also be implemented as an independent embodiment, and the order of the steps in a certain embodiment can be arbitrarily exchanged. In addition, the optional implementation methods in a certain embodiment can be arbitrarily combined; in addition, the embodiments can be arbitrarily combined, for example, some or all of the steps of different embodiments can be arbitrarily combined, and a certain embodiment can be arbitrarily combined with the optional implementation methods of other embodiments.

In each embodiment of the present disclosure, unless otherwise specified or there is a logical conflict, the terms and/or descriptions between the embodiments are consistent and can be referenced to each other, and the technical features in different embodiments can be combined to form a new embodiment based on their internal logical relationships.

The terms used in the embodiments of the present disclosure are only for the purpose of describing specific embodiments and are not intended to limit the present disclosure.

In the embodiments of the present disclosure, unless otherwise specified, elements expressed in the singular form, such as "a", "an", "the", "above", "said", "aforementioned", "this", etc., may mean "one and only one", or "one or more", "at least one", etc. For example, when using articles such as "a", "an", "the" in English in translation, the noun after the article may be understood as a singular expression or a plural expression.

In the embodiments of the present disclosure, “plurality” refers to two or more.

In some embodiments, "at least one", "one or more", The terms "a plurality of", "multiple" and the like are interchangeable.

In some embodiments, "at least one of A and B", "A and/or B", "A in one case, B in another case", "in response to one case A, in response to another case B", etc., may include the following technical solutions according to the situation: in some embodiments, A (A is executed independently of B); in some embodiments, B (B is executed independently of A); in some embodiments, execution is selected from A and B (A and B are selectively executed); in some embodiments, A and B (both A and B are executed). When there are more branches such as A, B, C, etc., the above is also similar.

In some embodiments, the recording method of "A or B" may include the following technical solutions according to the situation: in some embodiments, A (A is executed independently of B); in some embodiments, B (B is executed independently of A); in some embodiments, execution is selected from A and B (A and B are selectively executed). When there are more branches such as A, B, C, etc., the above is also similar.

The prefixes such as "first" and "second" in the embodiments of the present disclosure are only used to distinguish different description objects, and do not constitute restrictions on the position, order, priority, quantity or content of the description objects. The statement of the description object refers to the description in the context of the claims or embodiments, and should not constitute unnecessary restrictions due to the use of prefixes. For example, if the description object is a "field", the ordinal number before the "field" in the "first field" and the "second field" does not limit the position or order between the "fields", and the "first" and "second" do not limit whether the "fields" they modify are in the same message, nor do they limit the order of the "first field" and the "second field". For another example, if the description object is a "level", the ordinal number before the "level" in the "first level" and the "second level" does not limit the priority between the "levels". For another example, the number of description objects is not limited by the ordinal number, and can be one or more. Taking the "first device" as an example, the number of "devices" can be one or more. In addition, the objects modified by different prefixes may be the same or different. For example, if the description object is "device", then the "first device" and the "second device" may be the same device or different devices, and their types may be the same or different. For another example, if the description object is "information", then the "first information" and the "second information" may be the same information or different information, and their contents may be the same or different.

In some embodiments, “including A”, “comprising A”, “used to indicate A”, and “carrying A” can be interpreted as directly carrying A or indirectly indicating A.

In some embodiments, terms such as "time/frequency", "time/frequency domain", etc. refer to the time domain and/or the frequency domain.

In some embodiments, terms such as "in response to ...", "in response to determining ...", "in the case of ...", "at the time of ...", "when ...", "if ...", "if ...", etc. can be used interchangeably.

In some embodiments, terms such as "greater than", "greater than or equal to", "not less than", "more than", "more than or equal to", "not less than", "higher than", "higher than or equal to", "not lower than", and "above" can be replaced with each other, and terms such as "less than", "less than or equal to", "not greater than", "less than", "less than or equal to", "no more than", "lower than", "lower than or equal to", "not higher than", and "below" can be replaced with each other.

In some embodiments, devices and equipment may be interpreted as physical or virtual, and their names are not limited to the names recorded in the embodiments. In some cases, they may also be understood as "equipment", "device", "circuit", "network element", "node", "function", "unit", "section", "system", "network", "chip", "chip system", "entity", "subject", etc.

In some embodiments, "network" can be interpreted as devices included in the network, such as access network equipment, core network equipment, etc.

In some embodiments, "access network device (AN device)" may also be referred to as "radio access network device (RAN device)", "base station (BS)", "radio base station (radio base station)", "fixed station (fixed station)", and in some embodiments may also be understood as "node (node)", "access point (access point)", "transmission point (TP)", "reception point (RP)", "transmission and/or reception point (transmission/reception point)" point, TRP)", "panel", "antenna panel", "antenna array", "cell", "macro cell", "small cell", "femto cell", "pico cell", "sector", "cell group", "serving cell", "carrier", "component carrier", "bandwidth part (BWP)", etc.

In some embodiments, the term "terminal" or "terminal device" may be referred to as "user equipment (terminal)", "user terminal (user terminal)", "mobile station (MS)", "mobile terminal (MT)", subscriber station, mobile unit, subscriber unit, wireless unit, remote unit, mobile device, wireless device, wireless communication device, remote device, mobile subscriber station, access terminal, mobile terminal, wireless terminal, remote terminal, handset, user agent, mobile client, client, etc.

In some embodiments, acquisition of data, information, etc. may comply with the laws and regulations of the country where the data is obtained.

In some embodiments, data, information, etc. may be obtained with the user's consent.

In addition, each element, each row, or each column in the table of the embodiments of the present disclosure may be implemented as an independent embodiment, and the combination of any elements, any rows, and any columns may also be implemented as an independent embodiment.

FIG1 is a schematic diagram of the architecture of a communication system according to an embodiment of the present disclosure. As shown in FIG1 , the method provided in the embodiment of the present disclosure may be applied to a communication system 100, and the communication system may include a first device 101 and a second device 102. It should be noted that the communication system 100 may also include other devices, and the present disclosure does not limit the devices included in the communication system 100.

In some embodiments, the first device 101 is a terminal. Alternatively, the first device includes a terminal and a server corresponding to the terminal, and the terminal can obtain data from the server. Alternatively, the first device includes a terminal and a third-party server, and the terminal is connected to the third-party server wirelessly. Alternatively, the first device includes a terminal, a server corresponding to the terminal, and a third-party server, and the terminal is connected to the server corresponding to the terminal and the third-party server wirelessly.

In some embodiments, the second device 102 is a network device. Alternatively, the second device includes a network device and a core network, and the network device is connected to the core network. Alternatively, the second device includes a network device and a server corresponding to the network device, and the network device performs data transmission with the server corresponding to the network device.

In some embodiments, the terminal includes, for example, a mobile phone, a wearable device, an Internet of Things device, a car with communication function, a smart car, a tablet computer (Pad), a computer with wireless transceiver function, a virtual reality (VR) terminal device, an augmented reality (AR) terminal device, a wireless terminal device in industrial control (industrial control), a wireless terminal device in self-driving, a wireless terminal device in remote medical surgery, a wireless terminal device in a smart grid (smart grid), a wireless terminal device in transportation safety (transportation safety), a wireless terminal device in a smart city (smart city), and at least one of a wireless terminal device in a smart home (smart home), but is not limited to these.

In some embodiments, the network device may include at least one of an access network device and a core network device.

In some embodiments, the access network device is, for example, a node or device that accesses the terminal to the wireless network. The access network device may include an evolved NodeB (eNB), a next generation evolved NodeB (ng-eNB), and a next generation evolved NodeB (ng-eNB) in a 5G communication system. At least one of next generation NodeB (gNB), node B (NB), home node B (HNB), home evolved nodeB (HeNB), wireless backhaul equipment, radio network controller (RNC), base station controller (BSC), base transceiver station (BTS), base band unit (BBU), mobile switching center, base station in 6G communication system, open RAN, cloud RAN, base station in other communication systems, and access node in Wi-Fi system, but not limited thereto.

In some embodiments, the technical solution of the present disclosure may be applicable to the Open RAN architecture. In this case, the interfaces between access network devices or within access network devices involved in the embodiments of the present disclosure may become internal interfaces of Open RAN, and the processes and information interactions between these internal interfaces may be implemented through software or programs.

In some embodiments, the access network device may be composed of a centralized unit (central unit, CU) and a distributed unit (distributed unit, DU), wherein the CU may also be called a control unit (control unit). The CU-DU structure may be used to split the protocol layer of the access network device, with some functions of the protocol layer being centrally controlled by the CU, and the remaining part or all of the functions of the protocol layer being distributed in the DU, and the DU being centrally controlled by the CU, but not limited to this.

In some embodiments, the core network device may be a device including one or more network elements, or may be multiple devices or device groups, each including all or part of the one or more network elements. The network element may be virtual or physical. The core network may include, for example, at least one of the Evolved Packet Core (EPC), the 5G Core Network (5GCN), and the Next Generation Core (NGC).

It can be understood that the communication system described in the embodiment of the present disclosure is for the purpose of more clearly illustrating the technical solution of the embodiment of the present disclosure, and does not constitute a limitation on the technical solution proposed in the embodiment of the present disclosure. A person of ordinary skill in the art can know that with the evolution of the system architecture and the emergence of new business scenarios, the technical solution proposed in the embodiment of the present disclosure is also applicable to similar technical problems.

The following embodiments of the present disclosure may be applied to the communication system 100 shown in FIG1 , or part of the subject, but are not limited thereto. The subjects shown in FIG1 are examples, and the communication system may include all or part of the subjects in FIG1 , or may include other subjects other than FIG1 , and the number and form of the subjects are arbitrary, and the subjects may be physical or virtual, and the connection relationship between the subjects is an example, and the subjects may be connected or disconnected, and the connection may be in any manner, and may be a direct connection or an indirect connection, and may be a wired connection or a wireless connection.

The embodiments of the present disclosure may be applied to Long Term Evolution (LTE), LTE-Advanced (LTE-A), LTE-Beyond (LTE-B), SUPER 3G, IMT-Advanced, 4th generation mobile communication system (4G), 5th generation mobile communication system (5G), 5G new radio (NR), future radio access (FRA), new radio access technology (RAT), new radio (NR), new radio access (NX), future generation radio access (FX), Global System for Mobile communications (GSM (registered trademark)), CDMA2000, Ultra Mobile Broadband (UMB), IEEE 802.11 (Wi-Fi (registered trademark)), IEEE 802.16 (WiMAX (registered trademark)), IEEE 802.20, Ultra-WideBand (UWB), Bluetooth (Bluetooth (registered trademark)), Public Land Mobile Network (PLMN) network, Device-to-Device (D2D) system, Machine-to-Machine (M2M) system, Internet of Things (IoT) system, Vehicle-to-Everything (Vehicle-to-Everything) system. V2X), systems using other processing methods, and next-generation systems based on them. In addition, multiple systems can also be combined (for example, a combination of LTE or LTE-A and 5G, etc.) for application.

FIG2 is an interactive schematic diagram of a processing method according to an embodiment of the present disclosure. As shown in FIG2 , an embodiment of the present disclosure relates to a processing method, and the method includes:

Step S2101: the second device sends the first information.

In some embodiments, the first information is used by the first device to train a first decoding model so that the first decoding model has the ability to decode information of the same type as the first information.

In some embodiments, the first information includes first input data and first output data.

Optionally, the first input data includes a compressed measurement result; and the first output data includes an original measurement result or a second measurement result restored after the second decoding model decodes the first input data.

Optionally, the measurement result refers to a CSI measurement result. Or, it may be other measurement results.

For example, during the model application process, after the first device obtains the measurement result, it compresses the measurement result and sends the compressed measurement result to the second device, and the second device decodes the compressed measurement result using the second decoding model to obtain the second measurement result restored after decoding. Optionally, the compressed measurement result can also be understood as the first measurement result.

For another example, after obtaining the measurement result, the first device compresses the measurement result, and sends the compressed measurement result and the original measurement result to the second device.

In some embodiments, the second device sends the first information to the first device.

In some embodiments, the name of the first information is not limited, and it can be, for example, first training information, first indication information, etc.

In some embodiments, the first information further includes at least one of the following:

(1) Dataset identification.

In some embodiments, the data set identifier is used to indicate a data set. For example, the data set identifier is a data set ID.

(2) Whether the first input data is quantized data.

In some embodiments, the first input data refers to a compressed measurement result sent by the first device. The first device may further quantize the compressed measurement result and then send the compressed and quantized data.

Optionally, the first device may indicate whether the first input data is quantized data by a preset bit. Optionally, the preset bit is 1 bit. If the preset bit is 1, it indicates that it is quantized data. If the preset bit is 0, it indicates that it is not quantized data. Alternatively, the preset bit is 1 bit. If the preset bit is 0, it indicates that it is quantized data. If the preset bit is 1, it indicates that it is not quantized data.

(3) A quantization method for quantizing the first input data.

In some embodiments, if the first input data is quantized data, it is also necessary to indicate a quantization mode. Optionally, the quantization mode is scalar quantization, vector quantization of a codebook, or other types of quantization. For example, the scalar quantization is a 2-bit scalar quantization, a 4-bit scalar quantization, or other quantization. For example, the vector quantization of the codebook is a vector quantization with a codebook of (5, 10 bits), or other vector quantization.

(4) Scene information: The scene information is used to indicate the scene in which the first information is located.

In some embodiments, the scene information can be understood as an identifier of the above-mentioned first information, and the corresponding first information can be determined through the scene information.

(5) Configuration information, where the configuration information is used to indicate configuration of the first device to perform measurement.

In some embodiments, the configuration information includes at least one of the number of antenna ports, the number of frequency bands, and reporting overhead, or may also include other information, which is not limited in the embodiments of the present disclosure.

In some embodiments, the configuration information may be understood as an identifier of the above-mentioned first information, and the corresponding first information may be determined through the configuration information.

Step S2102: The first device obtains first input data and first output data.

In some embodiments, the first device receives first information, and the first information includes first input data and first output data. In some embodiments, the first device receives the first information sent by the second device, or it can also be understood that the first device receives the first input data and the first output data sent by the second device. Optionally, the second device is a base station, which means that the first device obtains the first input data and the first output data by receiving the first information sent by the base station. Optionally, the second device is a core network device, which means that the first device obtains the first input data and the first output data by receiving the first information sent by the core network device. Optionally, the second device is a network-side server, which means that the first device obtains the first input data and the first output data by receiving the first information sent by the network-side server. Optionally, the second device is a third-party server, which means that the first device obtains the first input data and the first output data by receiving the first information sent by the third-party server.

In some embodiments, the first device receives the first information sent by the second device, or it can also be understood that the first device receives the first input data and the first output data sent by other first devices. Optionally, the other first device is a UE side server. In this case, the first output data is the original measurement result.

Step S2103: The first device trains the first decoding model based on the first input data and the first output data to obtain a trained first decoding model.

In some embodiments, the first device inputs the first input data into the first decoding model to obtain decoded data after decoding, and then adjusts the parameters of the first decoding model based on the difference between the decoded data after decoding and the corresponding first output data, so that the first decoding model has the ability to decode the input data to obtain the corresponding output data.

In some embodiments, since the first input data and the first output data are both provided by the second device, after the first decoding model is trained using the first input data and the second output data, the trained first decoding model has the same decoding capability as the second decoding model of the second device.

In some embodiments, the first input data is provided by the first device, and the first output data is provided by the second device. In this case, the first input data is obtained by encoding the first output data provided by the second device by the first device.

Step S2104: The first device obtains a similarity difference based on the trained first decoding model.

In some embodiments, the similarity difference is used to indicate the difference between the first decoding model and the second decoding model. Alternatively, it can also be understood that the similarity difference is used to indicate the similarity between the first decoding model and the second decoding model. For example, the lower the similarity difference, the closer the first decoding model and the second decoding model are, and vice versa, the less similar the first decoding model and the second decoding model are.

After the first decoding model is trained, the similarity difference value can be obtained according to the decoding capability of the first decoding model, and then based on the similarity difference value, it is determined whether the operating performance of the second decoding model of the second device is normal.

In some embodiments, at least one first training similarity is determined based on the trained first decoding model, at least one second training similarity is determined based on the second decoding model, and a similarity difference is determined based on the at least one first training similarity and the at least one second training similarity.

Among them, the first training similarity is used to indicate the similarity between the third measurement result obtained by decoding the first measurement result by the trained first decoding model and the original measurement result, and the second training similarity is used to indicate the similarity between the fourth measurement result obtained by decoding the first measurement result by the second decoding model and the original measurement result.

In the disclosed embodiment, the compressed data is decoded using the trained first decoding model to obtain at least one first training similarity, and the data processed by the first decoding model is decoded using the second coding model to obtain at least one corresponding second training similarity, and then the similarity difference is determined based on the obtained at least one first training similarity and at least one second training similarity.

In some embodiments, at least one difference between at least one first training similarity and a corresponding second training similarity is obtained, and an average value of the at least one difference is determined as the similarity difference.

Optionally, if the acquired first training similarity is one, the corresponding second training similarity is also one, and the difference between the first training similarity and the corresponding second training similarity is directly determined as the similarity difference.

It should be noted that, in order to ensure the accuracy of the obtained similarity difference in the embodiment of the present disclosure, the number of the first training similarities and the second training similarities is multiple, so as to obtain multiple differences between the multiple first training similarities and the corresponding second training similarities, and then determine the average of the multiple differences as the similarity difference.

In some embodiments, the variance or standard deviation of the difference is less than the difference threshold. The difference threshold is agreed upon by the communication protocol, or agreed upon by the first device, or set in other ways, which is not limited in the embodiments of the present disclosure.

It should be noted that the similarity in the embodiment of the present disclosure may be SGCS, GCS, NMSE or other values, which is not limited in the embodiment of the present disclosure.

In some embodiments, a fifth measurement result of a first device is encoded using a first encoding model to obtain a first measurement result, the first measurement result is decoded using a trained first decoding model to obtain a third measurement result, and a first training similarity is determined based on the fifth measurement result and the third measurement result.

In the embodiment of the present disclosure, the fifth measurement result refers to the original measurement result obtained by the first device through measurement, that is, the measurement result before encoding by the first device.

In an embodiment of the present disclosure, the first device includes a first encoding model and a first decoding model. After the first device performs measurement to obtain a fifth measurement result, the first encoding model is first used to encode the fifth measurement result to obtain a first measurement result, and then the first decoding model is used to decode the first measurement result to obtain a third measurement result, and then the similarity between the fifth measurement result and the third measurement result is obtained.

In some embodiments, a first encoding model is used to encode a fifth measurement result of the first device to obtain a first measurement result, a second decoding model is used to decode the first measurement result to obtain a fourth measurement result, and a second training similarity is determined based on the fifth measurement result and the fourth measurement result.

In an embodiment of the present disclosure, the second device includes a second decoding model. After the first device performs measurement to obtain a fifth measurement result, the first encoding model is first used to encode the fifth measurement result to obtain a first measurement result, and the first measurement result is sent to the second device. The second device then uses the second decoding model to decode the first measurement result to obtain a fourth measurement result, and then obtains the similarity between the fifth measurement result and the fourth measurement result.

It should be noted that the embodiment of the present disclosure is described by taking the first device and the second device both encoding the fifth measurement result as an example. In another embodiment, the first device further quantizes the fifth measurement result, and the second device further dequantizes it, which is described in detail below.

In some embodiments, a fifth measurement result of a first device is encoded using a first encoding model to obtain an intermediate measurement result, the intermediate measurement result is quantized to obtain a first measurement result, the first measurement result is decoded using a trained first decoding model to obtain a third measurement result, and a first training similarity is determined based on the fifth measurement result and the third measurement result.

It should be noted that, when training the first decoding model and the second decoding model, the dequantization process also needs to be considered to ensure that the recovered data meets the requirements.

In some embodiments, the operating performance refers to the process of using the first decoding model to monitor the operating conditions of the second decoding model after the model training is completed in steps S2101-S2104. In other words, it refers to the operating performance of the second decoding model when the subsequent step S2105 starts to be executed.

Step S2105: The first device determines whether the operating performance of the second decoding model of the second device is higher than the second similarity based on the first similarity and the similarity difference of the first decoding model of the first device.

In some embodiments, the first decoding model and the second decoding model are used to decode the first measurement result of the first device.

In some embodiments, determining whether the operating performance of the second decoding model of the second device is higher than the second similarity can also be understood as determining whether the second decoding model of the second device operates normally, or, it can also be understood as determining whether the operation of the second decoding model of the second device meets the requirements, or, it can also be understood as determining whether the decoding accuracy of the second decoding model of the second device meets the requirements, etc.

In some embodiments, the sum of the first similarity and the similarity difference is obtained, and the sum is equal to or greater than the second similarity, and it is determined that the operating performance of the second decoding model is higher than the second similarity; or, the sum is less than the second similarity, and it is determined that the operating performance of the second decoding model is lower than the second similarity.

Optionally, due to the difference in operating conditions between the first device and the second device, there is also a difference in accuracy between the first decoding model of the first device and the second decoding model of the second device. Therefore, by comparing the sum of the first similarity and the similarity difference with the second similarity, if the sum of the first similarity and the similarity difference is stable, it also indicates that the second decoding model of the second device is operating stably.

The process of this application is described below by way of examples.

The first one:

1. The second device sends the first information to the first device in an offline manner.

Optionally, the first information includes: first input data, first output data, the quantization method is 2-bit scalar quantization, the scenario is the Ums scenario, the number of antenna ports is M, the reporting method is broadband reporting, and the output payload is Y.

2. The first device trains the first decoding model in an offline manner according to the received first information.

3. The first device reports the identifier of the first decoding model supported by itself through capability reporting.

4. The second device sends a downlink reference signal to the first device for channel measurement.

5. The first device performs channel measurement according to the downlink reference signal to obtain a fifth measurement result, compresses the fifth measurement result using the first coding model to obtain a first measurement result, and sends the first measurement result to the second device.

6. The second device decodes the first measurement result using the second decoding model to obtain a fourth measurement result.

The first device decodes the first measurement result using the first decoding model to obtain a third measurement result.

7. The second device sends the fourth measurement result to the first device, and the first device determines a first training similarity according to the fifth measurement result and the third measurement result, and determines a second training similarity according to the fifth measurement result and the fourth measurement result.

8. The first device determines a similarity difference according to the first training similarity and the second training similarity.

9. The first device supervises the second decoding model according to the similarity difference and the first similarity obtained by the trained first decoding model.

Second type:

1. The second device sends the first information to the first device in an offline manner.

Optionally, the first information includes: first input data, first output data, quantization method is 2-bit scalar quantization, scenario is Ums scenario, the number of antenna ports is M, reporting method is broadband reporting, output payload is Y, the first device trains the first decoding model in an offline manner, and the first device reports through capability information, reporting the number of antenna ports of the supported first decoding model, frequency domain granularity information reported by CSI, and CSI output payload information.

2. The first device trains the first decoding model in an offline manner according to the received first information.

3. When the original measurement results of the first device and the second device are consistent, the original measurement results are encoded and decoded in an offline manner, and the first training similarity is determined according to the restored measurement results and the original measurement results.

4. When the original measurement results of the first device and the second device are consistent, the original measurement results are encoded and decoded in an offline manner, and a second training similarity is determined according to the restored measurement results and the original measurement results.

5. The first device performs channel measurement according to the downlink reference signal to obtain a fifth measurement result, compresses the fifth measurement result using the first coding model to obtain a first measurement result, and sends the first measurement result to the second device.

6. The first device determines a similarity difference according to the first training similarity and the second training similarity.

7. The first device supervises the second decoding model according to the similarity difference and the first similarity obtained by the trained first decoding model to obtain a second similarity of the second device.

8. The first device reports an identifier and a similarity difference value of a supported first decoding model.

9. The second device determines the accuracy of the second decoding model according to the first similarity and the similarity difference, and determines whether to trigger model switching.

It should be noted that the content included in the first information in the first or second embodiment above can also be replaced by:

The first information includes: first input data, first output data, quantization method is 2-bit scalar quantization, scenario is Ums scenario, the number of antenna ports is M, reporting method is subband reporting, frequency domain information of each subband and the number of subbands.

or,

The first information includes: first input data, first output data, quantization method is (5, 10 bit) vector quantization, scene is Ums scene, and output dimension is X.

or,

The first information includes: first input data, first output data, quantization method is (5, 10 bit) vector quantization, scene is Ums scene, and output payload is Y.

In some embodiments, the names of information, etc. are not limited to the names recorded in the embodiments, and terms such as "information", "message", "signal", "signaling", "report", "configuration", "indication", "instruction", "command", "channel", "parameter", "domain", "field", "symbol", "symbol", "code element", "codebook", "codeword", "codepoint", "bit", "data", "program", and "chip" can be used interchangeably.

In some embodiments, terms such as "uplink", "uplink", "physical uplink" can be interchangeable, and terms such as "downlink", "downlink", "physical downlink" can be interchangeable, and terms such as "side", "sidelink", "side communication", "sidelink communication", "direct connection", "direct link", "direct communication", "direct link communication" can be interchangeable.

In some embodiments, "obtain", "obtain", "get", "receive", "transmit", "bidirectional transmission", "send and/or receive" can be interchangeable, and can be interpreted as receiving from other entities, obtaining from protocols, obtaining from high levels, obtaining by self-processing, autonomous implementation, etc.

In some embodiments, terms such as "send", "transmit", "report", "send", "transmit", "bidirectional transmission", "send and/or receive" can be used interchangeably.

In some embodiments, terms such as "moment", "time point", "time", and "time position" can be interchangeable, and terms such as "duration", "period", "time window", "window", and "time" can be interchangeable.

In some embodiments, terms such as "certain", "preset", "preset", "set", "indicated", "some", "any", and "first" can be interchangeable, and "specific A", "preset A", "preset A", "set A", "indicated A", "some A", "any A", and "first A" can be interpreted as A pre-defined in a protocol, etc., or as A obtained through setting, configuration, or indication, etc., and can also be interpreted as specific A, some A, any A, or first A, etc., but is not limited to this.

The processing method involved in the embodiments of the present disclosure may include at least one of steps S2101 to S2105. For example, step S2101 may be implemented as an independent embodiment, step S2102 may be implemented as an independent embodiment, step S2103 may be implemented as an independent embodiment, step S2104 may be implemented as an independent embodiment, step S2105 may be implemented as an independent embodiment, steps S2101, step S2102, step S2103, step S2104, step S2105 may be implemented as independent embodiments, steps S2102, step S2103, step S2104, step S2105 may be implemented as independent embodiments, steps S2101 and step S2106 may be implemented as independent embodiments, but are not limited thereto.

In some embodiments, step S2102, step S2103, step S2104, and step S2105 are optional, and one or more of these steps may be omitted or replaced in different embodiments.

In some embodiments, step S2101, step S2103, step S2104, and step S2105 are optional, and one or more of these steps may be omitted or replaced in different embodiments.

In some embodiments, steps S2101, S2102, S2104, and S2105 are optional. One or more of these steps may be omitted or replaced.

In some embodiments, steps S2101, S2102, S2103, and S2105 are optional, and one or more of these steps may be omitted or replaced in different embodiments.

In some embodiments, reference may be made to other optional implementations recorded before or after the description corresponding to FIG. 2 .

FIG3A is a flow chart of a processing method according to an embodiment of the present disclosure, which is applied to a first device. As shown in FIG3A , an embodiment of the present disclosure relates to a processing method, which includes:

Step S3101: The first device obtains first input data and first output data.

The optional implementation of step S3101 can refer to the optional implementation of step S2102 in FIG. 2 and other related parts in the embodiment involved in FIG. 2 , which will not be described in detail here.

Step S3102: The first device trains a first decoding model based on the first input data and the first output data to obtain a trained first decoding model.

The optional implementation of step S3102 can refer to the optional implementation of step S2103 in FIG. 2 and other related parts in the embodiment involved in FIG. 2 , which will not be described in detail here.

Step S3103: The first device obtains a similarity difference based on the trained first decoding model.

The optional implementation of step S3103 can refer to the optional implementation of step S2104 in FIG. 2 and other related parts in the embodiment involved in FIG. 2 , which will not be described in detail here.

Step S3104: The first device determines whether the operating performance of the second decoding model of the second device is higher than the second similarity based on the first similarity and the similarity difference of the first decoding model of the first device.

The optional implementation of step S3104 can refer to the optional implementation of step S2105 in FIG. 2 and other related parts in the embodiment involved in FIG. 2 , which will not be described in detail here.

The processing method involved in the embodiment of the present disclosure may include at least one of step S3101 to step S3104. For example, step S3101 may be implemented as an independent embodiment, step S3102 may be implemented as an independent embodiment, step S3103 may be implemented as an independent embodiment, step S3104 may be implemented as an independent embodiment, or at least two steps may be combined, but are not limited thereto.

In some embodiments, step S3101 and step S3102 are optional, step S3101 and step S3103 are optional, step S3102 and step S3103 are optional, step S3101 and step S3104 are optional, step S3101 is optional, step S3102 is optional, step S3103 is optional, and step S3104 is optional. In different embodiments, one or more of these steps may be omitted or replaced. But it is not limited thereto.

FIG3B is a flow chart of a processing method according to an embodiment of the present disclosure, which is applied to a terminal. As shown in FIG3B , an embodiment of the present disclosure relates to a processing method, which includes:

Step S3201: The first device obtains first input data and first output data.

The optional implementation of step S3201 can refer to the optional implementation of step S2102 in FIG. 2 and other related parts in the embodiment involved in FIG. 2 , which will not be described in detail here.

Step S3202: The first device trains the first decoding model based on the first input data and the first output data to obtain a trained The first decoding model.

The optional implementation of step S3202 can refer to the optional implementation of step S2103 in FIG. 2 and other related parts in the embodiment involved in FIG. 2 , which will not be described in detail here.

Step S3103: The first device obtains a similarity difference based on the trained first decoding model.

The optional implementation of step S3103 can refer to the optional implementation of step S2104 in FIG. 2 and other related parts in the embodiment involved in FIG. 2 , which will not be described in detail here.

FIG4 is a flow chart of a processing method according to an embodiment of the present disclosure, which is applied to a second device. As shown in FIG4 , an embodiment of the present disclosure relates to a processing method, and the method includes:

Step S4101: the second device sends the first information.

The optional implementation of step S4101 can refer to step S2101 in FIG. 2 and other related parts of the embodiment involved in FIG. 2 , which will not be described in detail here.

In some embodiments, the first device receives the first information sent by the second device, but is not limited thereto, and may also receive the first information sent by other entities.

In some embodiments, the first device obtains first information specified by a protocol.

In some embodiments, the first device obtains the first information from a higher layer.

In some embodiments, the first device performs processing to obtain the first information.

In some embodiments, step S4101 is omitted, and the first device autonomously implements the function indicated by the first information, or the above function is default or by default.

In combination with some embodiments of the first aspect, in some embodiments, the sum of the first similarity and the similarity difference is equal to or greater than the second similarity, and the operating performance of the second decoding model is higher than the second similarity; or, the sum of the first similarity and the similarity difference is less than the second similarity, and the operating performance of the second decoding model is lower than the second similarity.

In conjunction with some embodiments of the first aspect, in some embodiments, the similarity difference is obtained based on the trained first decoding model;

The trained first decoding model is obtained by training the first decoding model based on the first input data and the first output data;

The first input data includes a compressed measurement result; the first output data includes an original measurement result or a second measurement result restored after the second decoding model decodes the first measurement result of the first device.

In combination with some embodiments of the first aspect, in some embodiments, obtaining the similarity difference based on the trained first decoding model includes:

The similarity difference is determined based on the at least one first training similarity and the at least one second training similarity;

The at least one second training similarity is determined based on the second decoding model, and the second training similarity is used to indicate the similarity between the fourth measurement result obtained by decoding the first measurement result by the second decoding model and the original measurement result; the at least one first training similarity is determined based on the trained first decoding model, and the first training similarity is used to indicate the similarity between the third measurement result obtained by decoding the first measurement result by the trained first decoding model and the original measurement result. Spend.

In conjunction with some embodiments of the first aspect, in some embodiments, the similarity difference is an average value of the at least one difference;

The at least one difference value is a difference value between at least one of the first training similarities and a corresponding second training similarity.

In combination with some embodiments of the first aspect, in some embodiments, the variance or standard deviation of the difference is less than a difference threshold.

In combination with some embodiments of the first aspect, in some embodiments, the first training similarity is determined based on the fifth measurement result and the third measurement result; the first measurement result is obtained by encoding the fifth measurement result of the first device using a first coding model;

The third measurement result is obtained by decoding the first measurement result using the trained first decoding model.

In combination with some embodiments of the first aspect, in some embodiments, the second training similarity is determined based on the fifth measurement result and the fourth measurement result;

The first measurement result is obtained by encoding a fifth measurement result of the first device using a first coding model;

The fourth measurement result is obtained by decoding the first measurement result using the second decoding model.

In combination with some embodiments of the first aspect, in some embodiments, a second device sends the first information to the first device, and the first information includes the first input data and the first output data.

In conjunction with some embodiments of the first aspect, in some embodiments, the first information further includes at least one of the following:

Dataset identifier;

whether the first input data is quantitative data;

a quantization method for quantizing the first input data;

Scene information, where the scene information is used to indicate a scene in which the first information is located;

Configuration information, where the configuration information is used to indicate a configuration for the first device to perform measurement.

FIG5 is a flow chart of a processing method according to an embodiment of the present disclosure. As shown in FIG5 , the embodiment of the present disclosure relates to a processing method, and the method includes:

Step S5101, the terminal side receives a dataset (V1, V2) and related information from the network side.

In some embodiments, V1 is the input of the proxy decoder, such as compressed CSI information.

In some embodiments, V2 is the label of the proxy decoder, such as the CSI information recovered after decompression, or the ground-truth CSI.

In some embodiments, the relevant information may be the quantization information of V1, such as whether the compressed CSI is quantized information; what quantization method is used, e.g., 2-bit scalar quantization, vector quantization with codebook (5, 10 bits). In some embodiments, the relevant information may also be the scene information corresponding to the dataset, CSI configuration information, e.g., the number of antenna ports, the number of subbands (partial bandwidth), and reporting overhead.

In step S5102, the terminal side performs proxy model training based on the dataset and information transmitted by the network side.

In step S5103, the terminal side uses an encoder and a proxy decoder to encode and decode the CSI, and calculates SGCS#1 (spectral graph conyolutional networks) based on the recovered CSI and the real CSI.

In some embodiments, there may be multiple SGCS#1.

Step S5104: The terminal side uses the encoder and the decoder on the network side to encode and decode the CSI, and calculates SGCS#2 based on the recovered CSI and the real CSI.

In some embodiments, SGCS#2 may be multiple

Step S5105: The terminal side calculates the difference gap between SGCS#1 and SGCS#2.

In some embodiments, the note gap value can be the average of multiple groups of values, and the variance/standard deviation of the multiple groups of gap values should be smaller than a fixed value.

In the training process of the above proxy model, steps 1 to 5 can be offline or over the air. For example, steps 1 to 5 are all completed offline. Another example is that steps 1 to 2 are offline training, and steps 3 to 5 are completed through air interface signaling interaction.

In order to implement steps 3 to 5 through the air interface, the NW needs to transfer multiple SGCS#2 to the UE. The transfer process can be triggered by the UE (UE request SGCS#2) or the base station.

In some embodiments, the present application is described by way of examples.

Step 1: NW sends the following information to UE in offline form:

Dataset (V1, V2), V1 is the input of proxy decoder, V2 is the label of proxy decoder

The quantization method of V1 is 2-bit uniform scalar quantization

Dataset is the corresponding dataset in the Uma scenario

The number of base station antenna ports corresponding to the Dataset is M, and the frequency domain granularity of CSI reporting is wideband report.

The output payload corresponding to Dataset is Y

Step 2: UE trains the proxy decoder model offline

Note: Starting from step 3 is the over the air/online method

Step 3: UE reports the supported proxy decoder model ID information through UE capability report

Step 4: The base station sends a downlink reference signal for channel measurement

Step 5: The UE performs channel measurement based on the downlink reference signal, inputs the measurement result into the encoder model, and reports the compressed channel information to the base station according to the configuration of the base station.

Step 6:

The base station receives the compressed channel information from the UE and uses the decoder to restore the channel information

UE uses proxy decoder to restore channel information

Step 7:

The base station feeds back the channel information restored by the decoder to the UE. The UE calculates SGCS#1 based on the real CSI or the encoder model input CSI and the channel information restored by the base station.

UE calculates SGCS#2 based on the real CSI or encoder model input CSI and the channel information restored by the proxy decoder

Step 8:

UE calculates the gap between SGCS#1 and SGCS#2; note that the gap value can be the average of multiple sets of values, and the variance/standard deviation of the gap values should be less than the fixed value

Step 9: During the model supervision process, UE estimates the decoding accuracy SGCS#1 of NW based on SGCS#2 and GAP value. That is, when SGCS#2 decreases, it is considered that SGCS#1 also decreases.

Embodiment 2: NW sends the following information to UE in offline form:

Dataset (V1, V2), V1 is the input of proxy decoder, V2 is the label of proxy decoder

The quantization method of V1 is 2-bit uniform scalar quantization

Dataset is the corresponding dataset in the Uma scenario

The number of base station antenna ports corresponding to the Dataset is M, and the frequency domain granularity of CSI reporting is wideband report.

The output payload corresponding to Dataset is Y

UE trains the proxy decoder model offline

UE reports the number of base station antenna ports of the supported proxy decoder model, the frequency domain granularity information reported by CSI, and the CSI output payload information through capability reporting.

Step 2: UE trains the proxy decoder model offline

Step 3: When the original CSI of UE and NW are consistent, the offline method uses encoder and proxy decoder to encode and decode the original CSI, and calculates SGCS#1 based on the recovered CSI and the real CSI;

Step 4: When the original CSI of UE and NW are consistent, the encoder and decoder are used to encode and decode the original CSI in offline mode, and SGCS#2 is calculated based on the recovered CSI and the real CSI.

Step 5:

UE calculates the difference gap between SGCS#1 and SGCS#2; note that the gap value can be the average of multiple sets of values, and the variance/standard deviation of the gap values should be less than the fixed value

Step 6: During the model supervision process, UE estimates the decoding accuracy SGCS#1 of NW based on SGCS#2 and GAP value.

Note: Starting from step 6 is the over the air/online method

Step 7: UE reports the supported proxy decoder model ID information and gap information through UE capability report

Step 8: During the model supervision process, the UE calculates SGCS#2 and reports it to the NW. The NW determines the situation of SGCS#1 based on the gap and SGCS#2, that is, the decoding accuracy of the decoder. If SGCS#2 is too low, the NW can trigger model switching or fallback.

Embodiments 3, 4, and 5 are different in the content of the information sent.

Embodiment 3: NW sends the following information to UE:

Dataset (V1, V2), V1 is the input of proxy decoder, V2 is the label of proxy decoder

The quantization method of V1 is 2-bit uniform scalar quantization

Dataset is the corresponding dataset in the Uma scenario

The number of base station antenna ports corresponding to the Dataset is M, and the frequency domain granularity of CSI reporting is subband reporting. Frequency domain information of each subband, and the number of subbands.

Embodiment 4: NW sends the following information to UE:

Dataset (V1, V2), V1 is the encoder output, V2 is the encoder label

The quantization method of V1 is (5, 10 bit) vector quantization, which quantizes every 5 output dimensions into 10 bits.

Dataset is the corresponding dataset in the Uma scenario

The output dimension of Dataset is X

·

Embodiment 5: NW sends the following information to UE:

Dataset (V1, V2), V1 is the encoder output, V2 is the encoder label

The quantization method of V1 is (5, 10 bit) vector quantization, which quantizes every 5 output dimensions into 10 bits.

Dataset is the corresponding dataset in the Uma scenario

The output payload corresponding to Dataset is Y

In the embodiments of the present disclosure, part or all of the steps and their optional implementations may be arbitrarily combined with part or all of the steps in other embodiments, or may be arbitrarily combined with optional implementations of other embodiments.

The embodiments of the present disclosure also propose a device for implementing any of the above methods, for example, a device is proposed, the above device includes a unit or module for implementing each step performed by the terminal in any of the above methods. For another example, another device is also proposed, including a unit or module for implementing each step performed by a network device (such as an access network device, a core network function node, a core network device, etc.) in any of the above methods.

It should be understood that the division of the units or modules in the above device is only a division of logical functions, which can be fully or partially integrated into one physical entity or physically separated in actual implementation. In addition, the units or modules in the device can be implemented in the form of a processor calling software: for example, the device includes a processor, the processor is connected to a memory, and instructions are stored in the memory. The processor calls the instructions stored in the memory to implement any of the above methods or implement the functions of the units or modules of the above device, wherein the processor is, for example, a general-purpose processor, such as a central processing unit (CPU) or a microprocessor, and the memory is a memory inside the device or a memory outside the device. Alternatively, the units or modules in the device may be implemented in the form of hardware circuits, and the functions of some or all of the units or modules may be implemented by designing the hardware circuits. The hardware circuits may be understood as one or more processors; for example, in one implementation, the hardware circuits are application-specific integrated circuits (ASICs), and the functions of some or all of the above units or modules may be implemented by designing the logical relationship of the components in the circuits; for another example, in another implementation, the hardware circuits may be implemented by programmable logic devices (PLDs), and Field Programmable Gate Arrays (FPGAs) may be used as an example, which may include a large number of logic gate circuits, and the connection relationship between the logic gate circuits may be configured by configuring the configuration files, thereby implementing the functions of some or all of the above units or modules. All units or modules of the above devices may be implemented in the form of software called by the processor, or in the form of hardware circuits, or in the form of software called by the processor, and the remaining part may be implemented in the form of hardware circuits.

In the embodiments of the present disclosure, the processor is a circuit with signal processing capability. In one implementation, the processor may be a circuit with instruction reading and execution capability, such as a central processing unit (CPU), a microprocessor, a graphics processor, or a processor. In another implementation, the processor can realize certain functions through the logical relationship of the hardware circuit, and the logical relationship of the above hardware circuit is fixed or reconfigurable, such as the processor is a hardware circuit implemented by an application-specific integrated circuit (ASIC) or a programmable logic device (PLD), such as an FPGA. In a reconfigurable hardware circuit, the processor loads the configuration document to implement the hardware circuit configuration, which can be understood as the process of the processor loading instructions to implement the functions of some or all of the above units or modules. In addition, it can also be a hardware circuit designed for artificial intelligence, which can be understood as an ASIC, such as a neural network processing unit (NPU), a tensor processing unit (TPU), a deep learning processing unit (DPU), etc.

FIG6A is a schematic diagram of the structure of the first device proposed in an embodiment of the present disclosure. As shown in FIG6A , the first device 6100 may include: at least one of a transceiver module 6101 and a processing module 6102. In some embodiments, the processing module 6102 is used to

Acquire first input data and first output data, wherein the first input data includes a compressed measurement result; the first output data includes an original measurement result or a second measurement result restored after a second decoding model decodes the first measurement result of the first device, and the second decoding model is set on the second device;

Training a first decoding model based on the first input data and the first output data to obtain a trained first decoding model, wherein the first decoding model is set in the first device;

Based on the trained first decoding model, a similarity difference is obtained, and the similarity difference is used to indicate the difference between the first decoding model and the second decoding model. Optionally, the above-mentioned transceiver module is used to perform at least one of the communication steps such as sending and/or receiving performed by the first device 6100 in any of the above methods, which will not be repeated here. Optionally, the above-mentioned processing module is used to perform at least one of the other steps performed by the first device 6100 in any of the above methods, which will not be repeated here.

Optionally, the processing module 6102 is used to execute at least one of the communication steps such as processing performed by the first device in any of the above methods, which will not be repeated here.

FIG6B is a schematic diagram of the structure of the second device proposed in an embodiment of the present disclosure. As shown in FIG6B , the second device 6200 may include: at least one of a transceiver module 6201, a processing module 6202, etc. In some embodiments, the transceiver module 6201 is used to send first input data and first output data, wherein the first input data includes a compressed measurement result; the first output data includes an original measurement result or a second measurement result restored after a second decoding model decodes the first measurement result of the first device, and the second decoding model is provided in the second device; the first input data and the first output data are used to train the first decoding model, and the first decoding model is provided in the first device; the first decoding model after training is used to obtain a similarity difference, and the similarity difference is used to indicate the difference between the first decoding model and the second decoding model. Optionally, the above-mentioned transceiver module is used to perform at least one of the communication steps such as sending and/or receiving performed by the second device 6200 in any of the above methods (such as step S2101 but not limited thereto), which will not be repeated here.

Optionally, the processing module 6202 is used to execute at least one of the communication steps such as processing performed by the second device in any of the above methods, which will not be repeated here.

In some embodiments, the transceiver module may include a sending module and/or a receiving module, and the sending module and the receiving module may be separate or integrated. Optionally, the transceiver module may be interchangeable with the transceiver.

In some embodiments, the processing module can be a module or include multiple submodules. Optionally, the multiple submodules respectively execute all or part of the steps required to be executed by the processing module. Optionally, the processing module can be replaced with the processor.

FIG7A is a schematic diagram of the structure of a communication device 7100 proposed in an embodiment of the present disclosure. The communication device 7100 may be a network device (e.g., an access network device, a core network device, etc.), or a terminal (e.g., a user device, etc.), or a chip, a chip system, or a processor that supports a network device to implement any of the above methods, or a chip, a chip system, or a processor that supports a terminal to implement any of the above methods. The communication device 7100 may be used to implement the method described in the above method embodiment, and the details may refer to the description in the above method embodiment.

As shown in FIG. 7A , the communication device 7100 includes one or more processors 7101. The processor 7101 may be a general-purpose processor or a dedicated processor, for example, a baseband processor or a central processing unit. The baseband processor may be used to process the communication protocol and the communication data, and the central processing unit may be used to control the communication device (such as a base station, a baseband chip, a terminal device, a terminal device chip, a DU or a CU, etc.), execute a program, and process the data of the program. The communication device 7100 is used to execute any of the above methods.

In some embodiments, the communication device 7100 further includes one or more memories 7102 for storing instructions. Optionally, all or part of the memory 7102 may also be outside the communication device 7100.

In some embodiments, the communication device 7100 further includes one or more transceivers 7103. When the communication device 7100 includes one or more transceivers 7103, the transceiver 7103 performs at least one of the communication steps such as sending and/or receiving in the above method (for example, step S2101, step S2102, step S2103, step S2104, but not limited thereto).

In some embodiments, the transceiver may include a receiver and/or a transmitter, and the receiver and the transmitter may be separate or integrated. Optionally, the terms such as transceiver, transceiver unit, transceiver, transceiver circuit, etc. may be replaced with each other, the terms such as transmitter, transmission unit, transmitter, transmission circuit, etc. may be replaced with each other, and the terms such as receiver, receiving unit, receiver, receiving circuit, etc. may be replaced with each other.

In some embodiments, the communication device 7100 may include one or more interface circuits 7104. Optionally, the interface circuit 7104 is connected to the memory 7102, and the interface circuit 7104 may be used to receive signals from the memory 7102 or other devices, and may be used to send signals to the memory 7102 or other devices. For example, the interface circuit 7104 may read instructions stored in the memory 7102 and send the instructions to the processor 7101.

The communication device 7100 described in the above embodiments may be a network device or a terminal, but the scope of the communication device 7100 described in the present disclosure is not limited thereto, and the structure of the communication device 7100 may not be limited by FIG. 7A. The communication device may be an independent device or may be part of a larger device. For example, the communication device may be: 1) an independent integrated circuit IC, or a chip, or a chip system or subsystem; (2) a collection of one or more ICs, optionally, the above IC collection may also include a storage component for storing data and programs; (3) an ASIC, such as a modem; (4) a module that can be embedded in other devices; (5) a receiver, a terminal device, an intelligent terminal device, a cellular phone, a wireless device, a handheld device, a mobile unit, a vehicle-mounted device, a network device, a cloud device, an artificial intelligence device, etc.; (6) others, etc.

7B is a schematic diagram of the structure of a chip 7200 provided in an embodiment of the present disclosure. In the case where the communication device 7100 may be a chip or a chip system, reference may be made to the schematic diagram of the structure of the chip 7200 shown in FIG. 7B , but the present disclosure is not limited thereto.

The chip 7200 includes one or more processors 7201, and the chip 7200 is used to execute any of the above methods.

In some embodiments, the chip 7200 further includes one or more interface circuits 7202. Optionally, the interface circuit 7202 is connected to the memory 7203. The interface circuit 7202 can be used to receive signals from the memory 7203 or other devices, and the interface circuit 7202 can be used to send signals to the memory 7203 or other devices. For example, the interface circuit 7202 can read instructions stored in the memory 7203 and send the instructions to the processor 7201.

In some embodiments, the interface circuit 7202 performs at least one of the communication steps such as sending and/or receiving in the above method, and the processor 7201 performs at least one of the other steps.

In some embodiments, terms such as interface circuit, interface, transceiver pin, and transceiver may be used interchangeably.

In some embodiments, the chip 7200 further includes one or more memories 7203 for storing instructions. Optionally, all or part of the memory 7203 may be outside the chip 7200.

The present disclosure also proposes a storage medium, on which instructions are stored, and when the instructions are executed on the communication device 7100, the communication device 7100 executes any of the above methods. Optionally, the storage medium is an electronic storage medium. Optionally, the storage medium is a computer-readable storage medium, but is not limited to this, and it can also be a storage medium readable by other devices. Optionally, the storage medium can be a non-transitory storage medium, but is not limited to this, and it can also be a temporary storage medium.

The present disclosure also proposes a program product, which, when executed by the communication device 7100, enables the communication device 7100 to execute any of the above methods. Optionally, the program product is a computer program product.

The present disclosure also proposes a computer program, which, when executed on a computer, causes the computer to execute any one of the above methods.

Claims (21)

A processing method, characterized in that the method further comprises: Acquire first input data and first output data, wherein the first input data includes a compressed measurement result; the first output data includes an original measurement result or a second measurement result restored after a second decoding model decodes the first measurement result of the first device, and the second decoding model is set on the second device; Training a first decoding model based on the first input data and the first output data to obtain a trained first decoding model, wherein the first decoding model is set in the first device; Based on the trained first decoding model, a similarity difference value is obtained, where the similarity difference value is used to indicate the difference between the first decoding model and the second decoding model. The method according to claim 1, characterized in that the obtaining the similarity difference value based on the trained first decoding model comprises: Determine at least one first training similarity based on the trained first decoding model, where the first training similarity is used to indicate the similarity between a third measurement result obtained by decoding the first measurement result by the trained first decoding model and the original measurement result; Determine at least one second training similarity based on the second decoding model, where the second training similarity is used to indicate a similarity between a fourth measurement result obtained by decoding the first measurement result by the second decoding model and an original measurement result; The similarity difference is determined based on the at least one first training similarity and the at least one second training similarity. The method according to claim 2, characterized in that the determining the similarity difference based on the at least one first training similarity and the at least one second training similarity comprises: Obtaining at least one difference between at least one of the first training similarities and a corresponding second training similarity; An average value of the at least one difference value is determined as the similarity difference value. The method according to claim 3 is characterized in that the variance or standard deviation of the difference is less than the difference threshold. The method according to claim 2, characterized in that the determining at least one first training similarity based on the trained first decoding model comprises: Encoding the fifth measurement result of the first device by using a first coding model to obtain the first measurement result; Decoding the first measurement result using the trained first decoding model to obtain a third measurement result; The first training similarity is determined based on the fifth measurement result and the third measurement result. The method according to claim 2, characterized in that the determining at least one second training similarity based on the second decoding model comprises: Encoding the fifth measurement result of the first device by using a first coding model to obtain the first measurement result; Decoding the first measurement result using the second decoding model to obtain the fourth measurement result; The second training similarity is determined based on the fifth measurement result and the fourth measurement result. The method according to any one of claims 1 to 6, characterized in that the obtaining of the first input data and the first output data comprises: First information sent by a second device is received, where the first information includes the first input data and the first output data. The method according to claim 7, characterized in that the first information further includes at least one of the following: Dataset identifier; whether the first input data is quantitative data; a quantization method for quantizing the first input data; Scene information, where the scene information is used to indicate a scene in which the first information is located; Configuration information, where the configuration information is used to indicate a configuration for the first device to perform measurement. A processing method, characterized in that Sending first input data and first output data, wherein the first input data includes a compressed measurement result; the first output data includes an original measurement result or a second measurement result restored after a second decoding model decodes the first measurement result of the first device As a result, the second decoding model is set in the second device; The first input data and the first output data are used to train a first decoding model, and the first decoding model is set in the first device; The trained first decoding model is used to obtain a similarity difference value, and the similarity difference value is used to indicate the difference between the first decoding model and the second decoding model. The method according to claim 9, characterized in that the similarity difference is determined based on the at least one first training similarity and the at least one second training similarity; The at least one second training similarity is determined based on the second decoding model, and the second training similarity is used to indicate the similarity between the fourth measurement result obtained by decoding the first measurement result by the second decoding model and the original measurement result; the at least one first training similarity is determined based on the trained first decoding model, and the first training similarity is used to indicate the similarity between the third measurement result obtained by decoding the first measurement result by the trained first decoding model and the original measurement result. The method according to claim 10, characterized in that The similarity difference is an average value of the at least one difference; The at least one difference value is a difference value between at least one of the first training similarities and a corresponding second training similarity. The method according to claim 11 is characterized in that the variance or standard deviation of the difference is less than a difference threshold. The method according to claim 10, characterized in that The first training similarity is determined based on the fifth measurement result and the third measurement result; the first measurement result is obtained by encoding the fifth measurement result of the first device using a first coding model; The third measurement result is obtained by decoding the first measurement result using the trained first decoding model. The method according to claim 10, characterized in that The second training similarity is determined based on the fifth measurement result and the fourth measurement result; The first measurement result is obtained by encoding a fifth measurement result of the first device using a first coding model; The fourth measurement result is obtained by decoding the first measurement result using the second decoding model. The method according to any one of claims 9 to 14, characterized in that First information is sent to the first device, where the first information includes the first input data and the first output data. The method according to claim 15, characterized in that the first information further includes at least one of the following: Dataset identifier; whether the first input data is quantitative data; a quantization method for quantizing the first input data; Scene information, where the scene information is used to indicate a scene in which the first information is located; Configuration information, where the configuration information is used to indicate a configuration for the first device to perform measurement. A first device, characterized in that the first device comprises: a processing module, configured to obtain first input data and first output data, wherein the first input data includes a compressed measurement result; the first output data includes an original measurement result or a second measurement result restored after a second decoding model decodes the first measurement result of the first device, and the second decoding model is set on the second device; The processing module is further used to train a first decoding model based on the first input data and the first output data to obtain a trained first decoding model, wherein the first decoding model is set in the first device; The processing module is further used to obtain a similarity difference value based on the trained first decoding model, where the similarity difference value is used to indicate the difference between the first decoding model and the second decoding model. A second device, characterized in that the second device comprises: The transceiver module is used to send first input data and first output data, wherein the first input data includes a compressed measurement result; the first output data includes an original measurement result or a first measurement result of the first device decoded by a second decoding model and restored a second measurement result, wherein the second decoding model is set in a second device; The first input data and the first output data are used to train a first decoding model, and the first decoding model is set in the first device; The trained first decoding model is used to obtain a similarity difference value, and the similarity difference value is used to indicate the difference between the first decoding model and the second decoding model. A device, characterized in that the device comprises: one or more processors; Wherein, the device is used to execute the model operation performance determination method described in any one of claims 1 to 8 or 9 to 16. A communication system, characterized in that it includes a first device and a second device, wherein the first device is configured to implement the processing method described in any one of claims 1 to 8, and the second device is configured to implement the processing method described in any one of claims 9 to 16. A storage medium storing instructions, characterized in that when the instructions are executed on a communication device, the communication device executes the processing method as described in any one of claims 1 to 8, or executes the processing method as described in any one of claims 9 to 16.