Communication method, communication device, communication system, storage medium, and program product

Abstract

The embodiment of the invention relates to a communication method, communication equipment, a communication system, a storage medium and a program product. The communication method can be executed by a first node, and the method comprises the following steps: obtaining first information, the first information being used for indicating a first model; based on the first information, a model pairing relation is determined, the model pairing relation is used for indicating at least one model pair, each model pair comprises the first model and a second model, and the first model is obtained through training based on a third model and / or second information associated with the third model. According to the method and the device, the model pairing relationship is determined based on the first information, so that the second model paired with the first model is determined, the compatibility of bilateral models is ensured, and system performance loss caused by mismatching of the models is avoided.

Description

Communication methods, communication equipment, communication systems, storage media and software products

[0001] This disclosure relates to the field of communication technology, and in particular to a communication method, communication device, communication system, storage medium, and program product.

[0002] With the advancement of communication technology, artificial intelligence (AI) models, machine learning (ML) models, and other models have been introduced into communication systems. In channel state information (CSI) feedback scenarios, compressed CSI feedback can be achieved based on the CSI generation model on the terminal side, while CSI recovery can be achieved based on the CSI recovery model on the network device side.

[0003]

[0004] There may be multiple CSI generation models on the terminal side and multiple CSI recovery models on the network device side. On the one hand, how to achieve model pairing between the terminal side and the network device side is an urgent problem to be solved. On the other hand, even if pairing is completed, if one end retrains the model to obtain a new model, how to achieve pairing of the new model with the model on the other end is also an urgent problem to be solved.

[0005] This disclosure provides a communication method, communication device, communication system, storage medium, and program product.

[0006] According to a first aspect of the present disclosure, a communication method is proposed, executed by a first node, the method comprising: acquiring first information, the first information being used to indicate a first model; and determining a model pairing relationship based on the first information, the model pairing relationship being used to indicate at least one model pair, each model pair including a first model and a second model, the first model being trained based on a third model and / or second information associated with the third model.

[0007] According to a second aspect of the present disclosure, a communication method is proposed, executed by a second node, the method comprising: configuring first information for a first model, the first information being used to determine a model pairing relationship, the model pairing relationship being used to indicate at least one model pair, each model pair including a first model and a second model, the first model being trained based on a third model and / or second information associated with the third model.

[0008] According to a third aspect of the present disclosure, a communication method is proposed, executed by a third node, the method comprising: training a first model based on a third model and / or second information associated with the third model; acquiring first information for indicating the first model; sending the first information to the first node; the first information being used to trigger the first node to determine a model pairing relationship, the model pairing relationship being used to indicate at least one model pair, each model pair including a first model and a second model.

[0009] According to a fourth aspect of the present disclosure, a communication method is proposed, executed by a fifth node, the method comprising: receiving first information and first data processed by a first model; processing the first data using a second model paired with the first model; wherein the first information is used to indicate the first model, the second model is determined based on a model pairing relationship, the model pairing relationship is determined based on the first information, and the model pairing relationship is used to indicate at least one model pair, each model pair including the first model and the second model.

[0010] According to a fifth aspect of the present disclosure, a first node is proposed, comprising: a first transceiver module configured to acquire first information, the first information being used to indicate a first model; and a first processing module configured to determine a model pairing relationship based on the first information, the model pairing relationship being used to indicate at least one model pair, each model pair including a first model and a second model, the first model being trained based on a third model and / or second information associated with the third model.

[0011] According to a sixth aspect of the present disclosure, a second node is proposed, comprising: a second processing module configured to configure first information for a first model, the first information being used to determine a model pairing relationship, the model pairing relationship being used to indicate at least one model pair, each model pair including a first model and a second model, the first model being trained based on a third model and / or the second information associated with the third model.

[0012] According to a seventh aspect of the present disclosure, a third node is proposed, comprising: a third processing module configured to train a first model based on a third model and / or second information associated with the third model; and a third transceiver module configured to send first information to the first node; the first information being used to trigger the first node to determine a model pairing relationship, the model pairing relationship being used to indicate at least one model pair, each model pair including a first model and a second model.

[0013] According to an eighth aspect of the present disclosure, a fifth node is proposed, comprising: a fourth transceiver module configured to receive first information and first data processed by a first model; and a fourth processing module configured to process the first data using a second model paired with the first model; wherein the first information is used to indicate the first model, the second model is determined based on a model pairing relationship, the model pairing relationship is determined based on the first information, and the model pairing relationship is used to indicate at least one model pair, each model pair including the first model and the second model.

[0014] According to a ninth aspect of the present disclosure, a communication device is provided, comprising: one or more processors; wherein the communication device is configured to perform a communication method as described in any of the first to fourth aspects.

[0015] According to a tenth aspect of the present disclosure, a communication system is proposed, including a first node, a second node, a third node, a fourth node, and a fifth node; the first node is configured to implement the communication method as described in the first aspect; the second node is configured to implement the communication method as described in the second aspect; the third node is configured to implement the communication method as described in the third aspect; the fourth node is configured to send a third model and / or second information associated with the third model to the third node; and the fifth node is configured to implement the communication method as described in the fourth aspect.

[0016] According to an eleventh aspect of the present disclosure, a storage medium is provided that stores instructions, which, when executed on a communication device, cause the communication device to perform a communication method as described in any of the first to fourth aspects.

[0017] According to a twelfth aspect of the present disclosure, a computer program product is provided, including a computer program that, when executed by a processor, implements the communication method of any one of the first to fourth aspects.

[0018] According to a thirteenth aspect of the present disclosure, a computer program is provided that includes code, which, when executed by a processor, implements the communication method of any one of the first to fourth aspects.

[0019] According to a fourteenth aspect of the present disclosure, a chip or chip system is provided, the chip or chip system including processing circuitry configured to perform a communication method as described in any of the first to fourth aspects.

[0020] In this embodiment of the disclosure, the first node obtains first information for indicating the first model. The first model is trained based on the third model and / or the second information associated with the third model. The model pairing relationship is determined based on the first information, thereby determining the second model paired with the first model. This ensures the compatibility of the two-sided model and avoids system performance loss due to model mismatch.

[0021] To more clearly illustrate the technical solutions in the embodiments of this disclosure, the accompanying drawings required for the description of the embodiments are introduced below. The following drawings are only some embodiments of this disclosure and do not impose specific limitations on the protection scope of this disclosure.

[0022] Figure 1A is a schematic diagram of the architecture of a communication system according to an embodiment of the present disclosure.

[0023] Figure 1B is a schematic diagram illustrating CSI compression and recovery based on a bilateral model according to an embodiment of the present disclosure.

[0024] Figures 2A to 2D are interactive schematic diagrams of the communication method according to embodiments of the present disclosure.

[0025] Figures 3A to 3D are schematic flowcharts illustrating a terminal performing a communication method according to embodiments of the present disclosure.

[0026] Figures 3E to 3H are schematic flowcharts illustrating a network device performing a communication method according to embodiments of the present disclosure.

[0027] Figure 4A is a flowchart illustrating the first node performing a communication method according to an embodiment of the present disclosure.

[0028] Figure 4B is a flowchart illustrating the second node performing a communication method according to an embodiment of the present disclosure.

[0029] Figure 4C is a flowchart illustrating a third node performing a communication method according to an embodiment of the present disclosure.

[0030] Figure 4D is a schematic flowchart illustrating another method for a fifth node to perform communication according to an embodiment of the present disclosure.

[0031] Figure 4E is a schematic flowchart illustrating another method for a fourth node to perform communication according to an embodiment of the present disclosure.

[0032] Figures 5A to 5D are schematic diagrams of the structure of the first node, the second node, the third node, and the fifth node according to embodiments of the present disclosure.

[0033] Figure 6 is a schematic diagram of another structure of a communication device according to an embodiment of the present disclosure.

[0034] Figure 7 is a schematic diagram of a chip structure according to an embodiment of the present disclosure.

[0035] This disclosure provides a communication method, communication device, communication system, storage medium, and program product.

[0036] In a first aspect, embodiments of this disclosure propose a communication method executed by a first node, the method comprising: acquiring first information, the first information being used to indicate a first model; and determining a model pairing relationship based on the first information, the model pairing relationship being used to indicate at least one model pair, each model pair including a first model and a second model, the first model being trained based on a third model and / or second information associated with the third model.

[0037] In this embodiment of the disclosure, the first node obtains first information for indicating the first model, which is trained based on the third model. The first information determines the model pairing relationship, thereby determining the second model paired with the first model. This ensures the compatibility of the two-sided model and avoids system performance loss due to model mismatch.

[0038] In conjunction with some embodiments of the first aspect, in some embodiments, a first model is used to execute a first processing procedure, a second model is used to execute a second processing procedure, and the first processing procedure and the second processing procedure are inverse processes of each other.

[0039] In conjunction with some embodiments of the first aspect, in some embodiments, the first model and the third model are used to perform the first processing procedure, the third model being indicated by third information configured by the second node, and the first information being the same as the second information.

[0040] In this embodiment of the disclosure, by using second information indicating the third model to indicate the first model, the number of model identifiers can be reduced, the complexity of maintaining model identifiers in the system can be reduced, and the pairing efficiency can be improved.

[0041] In conjunction with some embodiments of the first aspect, in some embodiments, the second model and the third model are used to perform the second processing procedure, wherein the second model is the third model; or, the second model is trained based on the third model and / or second information associated with the third model.

[0042] In conjunction with some embodiments of the first aspect, in some embodiments, the first information includes at least one of the following: first indication information for indicating a first model; second indication information for indicating a first model and a second model; third indication information for indicating training data of the first model; fourth indication information for indicating a training session of the first model; fifth indication information for indicating the configuration of type association of training data of the first model; and sixth indication information for indicating the conditions for collecting training data of the first model.

[0043] In conjunction with some embodiments of the first aspect, in some embodiments, the second information includes at least one of the following: third information for indicating a third model; fourth information for indicating the model structure of the third model; fifth information for indicating the model parameters of the third model; and sixth information for indicating the training data of the third model.

[0044] In conjunction with some embodiments of the first aspect, in some embodiments, the first information is determined based on the third information indicating the third model and the first numerical calculation.

[0045] In this embodiment of the disclosure, the first information can be calculated and determined based on the third information and the first data. This can reduce the signaling interaction between the node that trained the first model and the node that configured the first information, thereby improving the efficiency of configuring the first information and matching the model.

[0046] In conjunction with some embodiments of the first aspect, in some embodiments, the first information is also used to trigger pairing of the first model.

[0047] Secondly, embodiments of this disclosure propose a communication method executed by a second node, the method comprising: configuring first information for a first model, the first information being used to determine a model pairing relationship, the model pairing relationship being used to indicate at least one model pair, each model pair including a first model and a second model, the first model being trained based on a third model and / or the second information associated with the third model.

[0048] In conjunction with some embodiments of the second aspect, in some embodiments, a first model is used to execute a first processing procedure, a second model is used to execute a second processing procedure, and the first processing procedure and the second processing procedure are inverse processes of each other.

[0049] In conjunction with some embodiments of the second aspect, in some embodiments, the first model and the third model are used to perform the first processing procedure, the third model being indicated by third information configured by the second node, and the first information being the same as the third information.

[0050] In conjunction with some embodiments of the second aspect, in some embodiments, the second model and the third model are used to perform the second processing procedure, and the second model is the third model; or, the second model is trained based on the third model and / or the second information associated with the third model.

[0051] In conjunction with some embodiments of the second aspect, in some embodiments, the first information includes at least one of the following: first indication information for indicating a first model; second indication information for indicating a first model and a second model; third indication information for indicating training data of the first model; fourth indication information for indicating a training session of the first model; fifth indication information for indicating the configuration of type association of training data of the first model; and sixth indication information for indicating the conditions for collecting training data of the first model.

[0052] In conjunction with some embodiments of the second aspect, in some embodiments, the second information includes at least one of the following: third information for indicating a third model; fourth information for indicating the model structure of the third model; fifth information for indicating the model parameters of the third model; and sixth information for indicating the training data of the third model.

[0053] In conjunction with some embodiments of the second aspect, in some embodiments, the first information is determined based on the third information indicating the third model and the first numerical calculation.

[0054] In conjunction with some embodiments of the second aspect, in some embodiments, the method further includes: sending first information to a third node, the third node being used to train a first model based on a third model.

[0055] In conjunction with some embodiments of the second aspect, in some embodiments, the method further includes: receiving a first message sent by a third node, the first message being used to request first information.

[0056] Thirdly, embodiments of this disclosure propose a communication method executed by a third node. The method includes: training a first model based on seventh information associated with the third model; acquiring first information for indicating the first model; sending the first information to the first node; the first information is used to trigger the first node to determine a model pairing relationship, the model pairing relationship is used to indicate at least one model pair, and each model pair includes a first model and a second model.

[0057] In conjunction with some embodiments of the third aspect, in some embodiments, the first model is used to execute the first processing procedure, the second model is used to execute the second processing procedure, and the first processing procedure and the second processing procedure are inverse processes of each other.

[0058] In conjunction with some embodiments of the third aspect, in some embodiments, the second model and the third model are used to perform the second processing procedure, wherein the second model is the third model; or, the second model is trained based on the third model and / or second information associated with the third model.

[0059] In conjunction with some embodiments of the third aspect, in some embodiments, the first model and the third model are used to perform the first processing procedure, the third model being indicated by third information configured by the second node, and the first information being the same as the third information.

[0060] In conjunction with some embodiments of the third aspect, in some embodiments, obtaining the first information associated with the first model includes: receiving the first information sent by the second node, wherein the second node is used to configure the first information for the first model.

[0061] In conjunction with some embodiments of the third aspect, in some embodiments, the first information and the seventh information are received simultaneously, the second node and the fourth node are the same device, and the fourth node is used to send the seventh information to the third node.

[0062] In conjunction with some embodiments of the third aspect, in some embodiments, the method further includes: sending a first message to a second node, the first message being used to request first information.

[0063] In conjunction with some embodiments of the third aspect, in some embodiments, the first information includes at least one of the following: first indication information for indicating a first model; second indication information for indicating a first model and a second model; third indication information for indicating training data of the first model; fourth indication information for indicating a training session of the first model; fifth indication information for indicating the configuration of type association of training data of the first model; and sixth indication information for indicating the conditions for collecting training data of the first model.

[0064] In conjunction with some embodiments of the third aspect, in some embodiments, the second information includes at least one of the following: third information for indicating a third model; fourth information for indicating the model structure of the third model; fifth information for indicating the model parameters of the third model; and sixth information for indicating the training data of the third model.

[0065] In conjunction with some embodiments of the third aspect, in some embodiments, the first information is determined based on the third information indicating the third model and the first numerical calculation.

[0066] Fourthly, embodiments of this disclosure propose a communication method executed by a fifth node, the method comprising: receiving first information and first data processed by a first model; processing the first data using a second model paired with the first model; wherein the first information is used to indicate the first model, the second model is determined based on a model pairing relationship, the model pairing relationship is determined based on the first information, the model pairing relationship is used to indicate at least one model pair, each model pair including the first model and the second model.

[0067] In conjunction with some embodiments of the fourth aspect, in some embodiments, a first model is used to perform a first processing procedure, a second model is used to perform a second processing procedure, the first processing procedure and the second processing procedure are inverse processes of each other, and the first model is trained based on a third model and / or second information associated with the third model.

[0068] In conjunction with some embodiments of the fourth aspect, in some embodiments, the first model and the third model are used to perform the first processing procedure, the third model being indicated by third information configured by the second node, and the first information being the same as the third information.

[0069] In conjunction with some embodiments of the fourth aspect, in some embodiments, the second model and the third model are used to perform the second processing procedure, wherein the second model is the third model; or, the second model is trained based on the third model and / or the second information associated with the third model.

[0070] In conjunction with some embodiments of the fourth aspect, in some embodiments, the first information includes at least one of the following: first indication information for indicating a first model; second indication information for indicating a first model and a second model; third indication information for indicating training data of the first model; fourth indication information for indicating a training session of the first model; fifth indication information for indicating the configuration of type association of training data of the first model; and sixth indication information for indicating the conditions for collecting training data of the first model.

[0071] In conjunction with some embodiments of the fourth aspect, in some embodiments, the second information includes at least one of the following: third information for indicating a third model; fourth information for indicating the model structure of the third model; fifth information for indicating the model parameters of the third model; and sixth information for indicating the training data of the third model.

[0072] In conjunction with some embodiments of the fourth aspect, in some embodiments the method further includes: receiving seventh information, the seventh information being used to indicate the second model.

[0073] In some embodiments, in conjunction with the fourth aspect, the method further includes: receiving eighth information, the eighth information being used to indicate a model pairing relationship; and determining a second model paired with the first model based on the model pairing relationship.

[0074] Fifthly, embodiments of this disclosure propose a first node, comprising: a first transceiver module configured to acquire first information, the first information being used to indicate a first model; and a first processing module configured to determine a model pairing relationship based on the first information, the model pairing relationship being used to indicate at least one model pair, each model pair including a first model and a second model, the first model being trained based on a third model and / or second information associated with the third model.

[0075] In conjunction with some embodiments of the fifth aspect, in some embodiments, a first model is used to execute a first processing procedure, a second model is used to execute a second processing procedure, and the first processing procedure and the second processing procedure are inverse processes of each other.

[0076] In conjunction with some embodiments of the fifth aspect, in some embodiments, the first model and the third model are used to perform the first processing procedure, the third model being indicated by third information configured by the second node, and the first information being the same as the third information.

[0077] In conjunction with some embodiments of the fifth aspect, in some embodiments, the second model and the third model are used to perform the second processing procedure, wherein the second model is the third model; or, the second model is trained based on the third model and / or second information associated with the third model.

[0078] In conjunction with some embodiments of the fifth aspect, in some embodiments, the first information includes at least one of the following: first indication information for indicating a first model; second indication information for indicating a first model and a second model; third indication information for indicating training data of the first model; fourth indication information for indicating a training session of the first model; fifth indication information for indicating the configuration of type association of training data of the first model; and sixth indication information for indicating the conditions for collecting training data of the first model.

[0079] In conjunction with some embodiments of the fifth aspect, in some embodiments, the second information includes at least one of the following: third information for indicating a third model; fourth information for indicating the model structure of the third model; fifth information for indicating the model parameters of the third model; and sixth information for indicating the training data of the third model.

[0080] In conjunction with some embodiments of the fifth aspect, in some embodiments, the first information is determined based on the third information indicating the third model and the first numerical calculation.

[0081] In conjunction with some embodiments of the fifth aspect, in some embodiments, the first information is also used to trigger pairing of the first model.

[0082] In a sixth aspect, embodiments of this disclosure propose a second node, comprising: a second processing module configured to configure first information for a first model, the first information being used to determine a model pairing relationship, the model pairing relationship being used to indicate at least one model pair, each model pair including a first model and a second model, the first model being trained based on a third model and / or the second information associated with the third model.

[0083] In conjunction with some embodiments of the sixth aspect, in some embodiments, a first model is used to execute a first processing procedure, a second model is used to execute a second processing procedure, and the first processing procedure and the second processing procedure are inverse processes of each other.

[0084] In conjunction with some embodiments of the sixth aspect, in some embodiments, the first model and the third model are used to perform the first processing procedure, the third model being indicated by third information configured by the second node, and the first information being the same as the third information.

[0085] In conjunction with some embodiments of the sixth aspect, in some embodiments, the second model and the third model are used to perform the second processing procedure, wherein the second model is the third model; or, the second model is trained based on the third model and / or second information associated with the third model.

[0086] In conjunction with some embodiments of the sixth aspect, in some embodiments, the first information includes at least one of the following: first indication information for indicating a first model; second indication information for indicating a first model and a second model; third indication information for indicating training data of the first model; fourth indication information for indicating a training session of the first model; fifth indication information for indicating the configuration of type association of training data of the first model; and sixth indication information for indicating the conditions for collecting training data of the first model.

[0087] In conjunction with some embodiments of the sixth aspect, in some embodiments, the second information includes at least one of the following: third information for indicating a third model; fourth information for indicating the model structure of the third model; fifth information for indicating the model parameters of the third model; and sixth information for indicating the training data of the third model.

[0088] In conjunction with some embodiments of the sixth aspect, in some embodiments, the first information is determined based on the third information indicating the third model and the first numerical calculation.

[0089] In conjunction with some embodiments of the sixth aspect, in some embodiments, the communication device further includes: a second transceiver model configured to send first information to a third node, the third node being used to train the first model based on the third model.

[0090] In conjunction with some embodiments of the second aspect, in some embodiments, the second transceiver module is further configured to receive a first message sent by a third node, the first message being used to request first information.

[0091] In a seventh aspect, embodiments of this disclosure propose a third node, comprising: a third processing module configured to train a first model based on seventh information associated with the third model; and a third transceiver module configured to send first information to the first node; the first information is used to trigger the first node to determine a model pairing relationship, the model pairing relationship being used to indicate at least one model pair, each model pair including a first model and a second model.

[0092] In conjunction with some embodiments of the seventh aspect, in some embodiments, a first model is used to execute a first processing procedure, a second model is used to execute a second processing procedure, and the first processing procedure and the second processing procedure are inverse processes of each other.

[0093] In conjunction with some embodiments of the seventh aspect, in some embodiments, the second model and the third model are used to perform the second processing procedure, wherein the second model is the third model; or, the second model is trained based on the third model and / or second information associated with the third model.

[0094] In conjunction with some embodiments of the seventh aspect, in some embodiments, the first model and the third model are used to perform the first processing procedure, the third model being indicated by third information configured by the second node, and the first information being the same as the third information.

[0095] In conjunction with some embodiments of the seventh aspect, in some embodiments, the third transceiver model is further configured to receive first information sent by the second node, the second node being used to configure the first information for the first model.

[0096] In conjunction with some embodiments of the seventh aspect, in some embodiments, the first information and the seventh information are received simultaneously, the second node and the fourth node are the same device, and the fourth node is used to send the seventh information to the third node.

[0097] In conjunction with some embodiments of the seventh aspect, in some embodiments, the third transceiver model is further configured to send a first message to the second node, the first message being used to request first information.

[0098] In conjunction with some embodiments of the seventh aspect, in some embodiments, the first information includes at least one of the following: first indication information for indicating a first model; second indication information for indicating a first model and a second model; third indication information for indicating training data of the first model; fourth indication information for indicating a training session of the first model; fifth indication information for indicating the configuration of type association of training data of the first model; and sixth indication information for indicating the conditions for collecting training data of the first model.

[0099] In conjunction with some embodiments of the seventh aspect, in some embodiments, the second information includes at least one of the following: third information for indicating a third model; fourth information for indicating the model structure of the third model; fifth information for indicating the model parameters of the third model; and sixth information for indicating the training data of the third model.

[0100] In conjunction with some embodiments of the seventh aspect, in some embodiments, the first information is determined based on the third information indicating the third model and the first numerical calculation.

[0101] Eighthly, embodiments of this disclosure provide a network device, comprising: a fourth transceiver module configured to receive first information and first data processed by a first model; and a fourth processing module configured to process the first data using a second model paired with the first model; wherein the first information is used to indicate the first model, the second model is determined based on a model pairing relationship, the model pairing relationship is determined based on the first information, and the model pairing relationship is used to indicate at least one model pair, each model pair including the first model and the second model.

[0102] In conjunction with some embodiments of the eighth aspect, in some embodiments, a first model is used to perform a first processing procedure, a second model is used to perform a second processing procedure, the first processing procedure and the second processing procedure are inverse processes of each other, and the first model is trained based on a third model and / or second information associated with the third model.

[0103] In conjunction with some embodiments of the eighth aspect, in some embodiments, the first model and the third model are used to perform the first processing procedure, the third model being indicated by third information configured by the second node, and the first information being the same as the third information.

[0104] In conjunction with some embodiments of the eighth aspect, in some embodiments, the second model and the third model are used to perform the second processing procedure, wherein the second model is the third model; or, the second model is trained based on the third model and / or second information associated with the third model.

[0105] In conjunction with some embodiments of the eighth aspect, in some embodiments, the first information includes at least one of the following: first indication information for indicating a first model; second indication information for indicating a first model and a second model; third indication information for indicating training data of the first model; fourth indication information for indicating a training session of the first model; fifth indication information for indicating the configuration of type association of training data of the first model; and sixth indication information for indicating the conditions for collecting training data of the first model.

[0106] In conjunction with some embodiments of the eighth aspect, in some embodiments, the second information includes at least one of the following: third information for indicating a third model; fourth information for indicating the model structure of the third model; fifth information for indicating the model parameters of the third model; and sixth information for indicating the training data of the third model.

[0107] In conjunction with some embodiments of the eighth aspect, in some embodiments, the fourth transceiver module is also configured to receive seventh information, which is used to indicate the second model.

[0108] In conjunction with some embodiments of the eighth aspect, in some embodiments, the fourth transceiver module is further configured to receive eighth information, which is used to indicate a model pairing relationship; and based on the model pairing relationship, to determine a second model paired with the first model.

[0109] In a ninth aspect, embodiments of this disclosure provide a communication device, comprising: one or more processors; wherein the communication device is configured to perform a communication method as described in any of the first to fourth aspects.

[0110] In a tenth aspect, embodiments of this disclosure provide a communication system comprising: a first node, a second node, a third node, a fourth node, and a fifth node; the first node is configured to implement the communication method as described in the first aspect; the second node is configured to implement the communication method as described in the second aspect; the third node is configured to implement the communication method as described in the third aspect; the fourth node is configured to send a third model and / or second information associated with the third model to the third node; and the fifth node is configured to implement the communication method as described in the fourth aspect.

[0111] Eleventhly, embodiments of this disclosure provide a storage medium storing instructions that, when executed on a communication device, cause the communication device to perform a communication method as described in any of the first to fourth aspects.

[0112] In a twelfth aspect, embodiments of this disclosure provide a program product that, when executed by a communication device, causes the communication device to perform a communication method as described in any of the first to fourth aspects.

[0113] In a thirteenth aspect, embodiments of this disclosure provide a computer program that, when run on a computer, causes the computer to perform the method as described in an optional implementation of any of the first to fourth aspects.

[0114] In a fourteenth aspect, embodiments of this disclosure provide a chip or chip system. The chip or chip system includes processing circuitry configured to perform the method described according to an optional implementation of any of the first to fourth aspects.

[0115] It is understood that the aforementioned communication devices, communication systems, storage media, program products, computer programs, chips, or chip systems are all used to execute the methods proposed in the embodiments of this disclosure. Therefore, the beneficial effects they can achieve can be referred to the beneficial effects in the corresponding methods, and will not be repeated here.

[0116] This disclosure provides a communication method, communication device, communication system, storage medium, and program product. In some embodiments, the terms "communication method" and "model pairing method," "method for determining pairing relationships," etc., can be used interchangeably, as can the terms "information processing system," "communication system," and "pairing system."

[0117] This disclosure is not exhaustive, but merely illustrative of some embodiments, and is not intended to limit the scope of protection of this disclosure. Unless otherwise specified, each step in a particular 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 particular embodiment can also be implemented as an independent embodiment, and the order of the steps in a particular embodiment can be arbitrarily interchanged. Furthermore, the optional implementation methods in a particular embodiment can be arbitrarily combined; moreover, the embodiments can be arbitrarily combined, for example, some or all steps of different embodiments can be arbitrarily combined, and a particular embodiment can be arbitrarily combined with the optional implementation methods of other embodiments.

[0118] In each of the disclosed embodiments, unless otherwise specified or in case of logical conflict, the terminology and / or descriptions of the embodiments are consistent and can be referenced by each other. Technical features in different embodiments can be combined to form new embodiments based on their inherent logical relationships.

[0119] The terminology used in the embodiments of this disclosure is for the purpose of describing particular embodiments only and is not intended to limit the scope of this disclosure.

[0120] In this embodiment of the disclosure, unless otherwise stated, elements expressed in the singular form, such as "a," "an," "the," "the," "the," "the," "the," "the," "this," etc., can mean "one and only one," or "one or more," "at least one," etc. For example, when using articles such as "a," "an," "the," etc. in translation, the noun following the article can be understood as either a singular expression or a plural expression.

[0121] In the embodiments disclosed herein, "multiple" refers to two or more.

[0122] In some embodiments, the terms “at least one of”, “one or more”, “a plurality of”, “multiple”, etc., may be used interchangeably.

[0123] In some embodiments, the notation "at least one of A and B", "A and / or B", "A in one case, B in another", "in response to one case A, in response to another case B", etc., may include the following technical solutions depending on the situation: in some embodiments, A (execute A regardless of B); in some embodiments, B (execute B regardless of A); in some embodiments, execution is selected from A and B (A and B are selectively executed); in some embodiments, both A and B are executed. The same applies when there are more branches such as A, B, C, etc.

[0124] In some embodiments, the notation "A or B" may include the following technical solutions, depending on the situation: in some embodiments, A (execution of A regardless of B); in some embodiments, B (execution of B regardless of A); in some embodiments, execution is selected from A and B (A and B are selectively executed). The same applies when there are more branches such as A, B, C, etc.

[0125] The prefixes "first," "second," etc., used in the embodiments of this disclosure are merely for distinguishing different descriptive objects and do not impose restrictions on the position, order, priority, quantity, or content of the descriptive objects. The description of the descriptive objects is found in the claims or the context of the embodiments, and the use of prefixes should not constitute unnecessary restrictions. For example, if the descriptive object is a "field," the ordinal numbers preceding "field" in "first field" and "second field" do not restrict the position or order of the "fields." "First" and "second" do not restrict whether the "fields" they modify are in the same message, nor do they restrict the order of "first field" and "second field." Similarly, if the descriptive object is a "level," the ordinal numbers preceding "level" in "first level" and "second level" do not restrict the priority between "levels." Furthermore, the number of descriptive objects is not limited by ordinal numbers and can be one or more. For example, in "first device," the number of "devices" can be one or more. Furthermore, the objects modified by different prefixes can be the same or different. For example, if the object being described is "device", then "first device" and "second device" can be the same device or different devices, and their types can be the same or different. Similarly, if the object being described is "information", then "first information" and "second information" can be the same information or different information, and their content can be the same or different.

[0126] In some embodiments, “including A,” “containing A,” “for indicating A,” and “carrying A” can be interpreted as directly carrying A or indirectly indicating A.

[0127] In some embodiments, the terms “in response to…”, “in response to determining…”, “in the case of…”, “when…”, “if…”, “if…”, etc., can be used interchangeably.

[0128] In some embodiments, the terms “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 used interchangeably, as can the terms “less than,” “less than or equal to,” “not greater than,” “less than,” “less than or equal to,” “not more than,” “lower than,” “lower than or equal to,” “not higher than,” and “below”.

[0129] In some embodiments, devices, etc., can be interpreted as physical or virtual, and their names are not limited to the names recorded in the embodiments. Terms such as “device”, “equipment”, “circuit”, “network element”, “node”, “function”, “unit”, “section”, “system”, “network”, “chip”, “chip system”, “entity”, and “subject” can be used interchangeably.

[0130] In some embodiments, "network" can be interpreted as devices included in a network (e.g., access network devices, core network devices, etc.).

[0131] In some embodiments, the terms "network devices", "access network device (AN device)", "radio access network device (RAN device)", "base station (BS)", "radio base station", "fixed station", "node", "access network node", "access point", "transmission point (TP)", "reception point (RP)", "transmission / reception point (TRP)", "panel", "antenna panel", "antenna array", "cell", "macro cell", "small cell", "femtocell", "pico cell", "sector", "cell group", "serving cell", "carrier", "component carrier", and "bandwidth part (BWP)" can be used interchangeably.

[0132] In some embodiments, the terms "terminal", "terminal device", "user equipment (UE)", "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", and "client" can be used interchangeably.

[0133] In some embodiments, access network devices, core network devices, or network devices can be replaced by terminals. For example, embodiments of this disclosure can also be applied to structures where communication between access network devices, core network devices, or network devices and terminals is replaced by communication between multiple terminals (e.g., device-to-device (D2D), vehicle-to-everything (V2X), etc.). In this case, the structure can also be configured such that the terminal has all or part of the functions of the access network device. Furthermore, terms such as "uplink" and "downlink" can be replaced with terms corresponding to communication between terminals (e.g., "sidelink"). For example, uplink channel, downlink channel, etc., can be replaced with sidelink channel, and uplink link, downlink, etc., can be replaced with sidelink link.

[0134] In some embodiments, the terminal may be replaced by an access network device, a core network device, or a network device. In this case, the access network device, core network device, or network device may also be configured to have all or some of the functions of the terminal.

[0135] In some embodiments, the acquisition of data, information, etc., may comply with the laws and regulations of the country where the location is situated.

[0136] In some embodiments, data, information, etc., may be obtained with the user's consent.

[0137] Furthermore, each element, each row, or each column in the table of this disclosure can be implemented as an independent embodiment, and any combination of any element, any row, or any column can also be implemented as an independent embodiment.

[0138] Figure 1A is a schematic diagram of an architecture of a communication system according to an embodiment of the present disclosure. As shown in Figure 1A, the communication system 100 includes: a first node 101, a second node 102, a third node 103, a fourth node 104, and a fifth node 105.

[0139] In some embodiments, the first node is used to perform model pairing.

[0140] In some embodiments, the first node is used to determine model pairing relationships.

[0141] In some embodiments, the name of the first node is not limited, and it may be, for example, "pairing node", "matching node", "associated node", etc.

[0142] In some embodiments, the second node is used to configure indication information.

[0143] In some embodiments, the indication information can be used to indicate a model, in which case the indication information is the model's indication information. Alternatively, the indication information can be used to indicate the model's training data, in which case the indication information is the training data's indication information. Alternatively, the indication information can be used to indicate a model pair, in which case the indication information is the model pair's indication information. Alternatively, the indication information can be used to indicate a model's training session, in which case the indication information is the training session's indication information. Alternatively, the indication information can be used to indicate the configuration associated with the type of training data, in which case the indication information is the configuration's indication information. Alternatively, the indication information can be used to indicate the conditions for collecting the training data, in which case the indication information is the condition's indication information.

[0144] In some embodiments, the second node may be configured with indication information for at least one of the following: model, model pair, training data, training session, configuration associated with training data type, and conditions for collecting training data.

[0145] In some embodiments, the indication information may be an identifier (ID).

[0146] In some embodiments, the name of the second node is not limited, and may be, for example, "configuration node", "assignment node", "ID assignment node", "ID configuration node", etc.

[0147] In some embodiments, the third node can be used to train the first model based on the third model.

[0148] In some embodiments, the third node can be used to train the first model based on the seventh information associated with the third model.

[0149] In some embodiments, the third node can be used to trigger model pairing.

[0150] In some embodiments, the name of the third node is not limited, and may be, for example, "trigger node", "initiator node", "request node", "pairing trigger node", "pairing initiator node", "pairing request node", "training node", etc.

[0151] In some embodiments, the fourth node can be used to send a third model;

[0152] In some embodiments, the fourth node can be used to send the seventh information associated with the third model.

[0153] In some embodiments, the name of the fourth node is not limited, and it may be, for example, "sending node", "transfer node", "model sending node", "model transfer node", etc.

[0154] In some embodiments, the fifth node can be used to find the model used from the model pairing relationship.

[0155] In some embodiments, the name of the fifth node is not limited, and it may be, for example, “model lookup node”, “model application node”, or “model use node”.

[0156] In some embodiments, the first node, second node, third node, fourth node, and fifth node may be a terminal or network device.

[0157] In some embodiments, the first node, second node, third node, fourth node and fifth node described above may be deployed in one device or in multiple devices, each device having the functions of one or more of the above nodes.

[0158] In one example, terminal A has a third node deployed, and network access device B has a first node, a second node, a fourth node, and a fifth node deployed.

[0159] In one example, terminal A has a second node, a fourth node, and a fifth node deployed, while access network device B has a first node and a third node deployed.

[0160] In one example, terminal A has a first node, a fourth node, and a fifth node deployed, while network access device B has a second node and a third node deployed.

[0161] In one example, terminal A is deployed with a fourth and fifth node, access network device B is deployed with a third node, and core network device C is deployed with a first and second node.

[0162] In one example, terminal A is deployed with a third node, access network device B is deployed with a fourth node, core network device C is deployed with a first node, core network device D is deployed with a second node, and core network device E is deployed with a fifth node.

[0163] In some embodiments, the terminal includes, but is not limited to, at least one of the following: mobile phone, wearable device, Internet of Things device, car with communication function, smart car, tablet computer, computer with wireless transceiver function, virtual reality (VR) terminal device, augmented reality (AR) terminal device, wireless terminal device in industrial control, wireless terminal device in self-driving, wireless terminal device in remote medical surgery, wireless terminal device in smart grid, wireless terminal device in transportation safety, wireless terminal device in smart city, and wireless terminal device in smart home.

[0164] In some embodiments, network devices may include access network devices and / or core network devices. Access network devices are, for example, nodes or devices that connect terminals to a wireless network. Access network devices may include, but are not limited to, at least one of the following: evolved NodeB (eNB), next-generation eNB (ng-eNB), next-generation NodeB (gNB), node B (NB), home node B (HNB), home evolved node B (HeNB), wireless backhaul device, radio network controller (RNC), base station controller (BSC), base transceiver station (BTS), base band unit (BBU), mobile switching center, base station in a 6G communication system, open RAN, cloud RAN, base station in other communication systems, and access node in a Wi-Fi system.

[0165] In some embodiments, the technical solutions of this disclosure can be applied to the Open RAN architecture. In this case, the interfaces between or within network devices involved in the embodiments of this disclosure can be transformed into internal interfaces of Open RAN. The processes and information interactions between these internal interfaces can be implemented by software or programs.

[0166] In some embodiments, the access network device may be composed of a central unit (CU) and a distributed unit (DU). The CU may also be called a control unit. The CU-DU structure can separate the protocol layer of the network device. Some of the protocol layer functions are centrally controlled by the CU, while the remaining part or all of the protocol layer functions are distributed in the DU and centrally controlled by the CU. However, this is not the only possibility.

[0167] In some embodiments, the core network equipment may be a single device including a first network element, or it may be multiple devices or a group of devices, each including a first network element. Network elements may be virtual or physical. The core network may include, for example, at least one of an Evolved Packet Core (EPC), a 5G Core Network (5GCN), or a Next Generation Core (NGC).

[0168] It is understood that the communication system described in this disclosure is for the purpose of more clearly illustrating the technical solutions of this disclosure, and does not constitute a limitation on the technical solutions provided in this disclosure. As those skilled in the art will know, with the evolution of system architecture and the emergence of new business scenarios, the technical solutions provided in this disclosure are also applicable to similar technical problems.

[0169] The following embodiments of this disclosure can be applied to the communication system 100 shown in FIG1A, or to some of the main bodies, but are not limited thereto. The main bodies shown in FIG1A are illustrative. The communication system may include all or some of the main bodies in FIG1A, or it may include other main bodies outside of FIG1A. The number and form of each main body are arbitrary. The connection relationship between the main bodies is illustrative. The main bodies may not be connected or may be connected. The connection can be in any way, it can be a direct connection or an indirect connection, it can be a wired connection or a wireless connection.

[0170] The embodiments disclosed herein can 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), CDMA2000, Ultra Mobile Broadband (UMB), IEEE 802.11 (Wi-Fi), IEEE 802.16 (WiMAX), and IEEE 802.20, Ultra-Wideband (UWB), Bluetooth (a registered trademark), Public Land Mobile Network (PLMN) networks, Device-to-Device (D2D) systems, Machine-to-Machine (M2M) systems, Internet of Things (IoT) systems, Vehicle-to-Everything (V2X) systems, systems utilizing other communication methods, and next-generation systems built upon them, etc. Furthermore, multiple systems can be combined (e.g., a combination of LTE or LTE-A with 5G).

[0171] The following is an explanation and interpretation of the terminology used in this disclosure.

[0172] Figure 1B is a schematic diagram of CSI compression and recovery based on a bilateral model provided in an embodiment of this disclosure. As shown in Figure 1B, the UE can compress the downlink channel information H and quantize it into a binary bit stream s through the CSI generation model and send it to the gNB. The gNB can recover H′, which is similar to the original downlink information, through the CSI recovery model.

[0173] In some embodiments, the methods for training the CSI generation model and the CSI recovery model include at least one of the following:

[0174] (1) The model is trained on one side (such as the terminal side or the network device side), and then the trained model is sent to the other side.

[0175] (2) Train the CSI generation model and the CSI recovery model separately on the terminal side and the network device side through joint training. Alternatively, after training part of the model on one end of the terminal side or the network side, train another part of the bilateral model on the other end, where the parameters of the model trained first are not updated.

[0176] (3) First, complete the training of the model on one side, and then send the training data or other auxiliary information to the other side to complete the training of the other part of the model.

[0177] In some embodiments, the following options have been proposed to reduce or mitigate the training complexity of bilateral models:

[0178] Method 1: Standardize the model structure and parameters.

[0179] Method 2: Dataset standardization.

[0180] Method 3: Standardize the model structure, and the model parameters are passed between the network device side and the terminal side.

[0181] Method 4: Standardize the data format, and transmit the data between the network device side and the terminal side.

[0182] Method 5: Standardize the model format, and the reference model is transmitted between the network device side and the terminal side.

[0183] In some embodiments, based on the model parameters passed to the terminal or the behavior executed by the terminal after modeling, methods 3 and 5 can be further divided into:

[0184] For method 3:

[0185] Method 3a: Receive the model parameters and train the model again to develop a different model.

[0186] Method 3b: The received model parameters are directly used for model inference.

[0187] For method 5:

[0188] Method 5a: Train the received model and redevelop a different model.

[0189] In some embodiments, depending on the content of the transmitted data, method 4 can be further divided into:

[0190] For method 4, the dataset is transmitted from the network device side to the terminal side, then

[0191] Method 4a: The dataset consists of target CSI and feedback CSI.

[0192] Method 4b: The dataset consists of the feedback CSI and the reconstructed target CSI.

[0193] Method 4c: The dataset consists of the target CSI, the feedback CSI, and the recovered target CSI.

[0194] There may be multiple CSI generation models on the terminal side and multiple CSI recovery models on the network device side. On the one hand, how to achieve model pairing between the terminal side and the network device side is an urgent problem to be solved. On the other hand, even if pairing is completed, if one end retrains the model to obtain a new model, how to achieve pairing of the new model with the model on the other end is also an urgent problem to be solved.

[0195] This disclosure provides a communication method, communication device, communication system, storage medium, and program product. It obtains first information for indicating a first model, which is trained based on a third model and / or second information associated with the third model. Based on the first information, it determines the model pairing relationship, thereby determining the second model paired with the first model. This ensures the compatibility of the two-sided models and avoids system performance loss due to model mismatch.

[0196] In some embodiments, a first model is used to execute a first processing procedure, and a second model is used to execute a second processing procedure, wherein the first processing procedure and the second processing procedure are inverse processes of each other.

[0197] In some embodiments, the first processing step may be an encoding process or a decoding process. If the first processing step is an encoding process, the second processing step is a decoding process. If the first processing step is a decoding process, the second processing step is an encoding process.

[0198] In some embodiments, the first processing step can be a compression process or a recovery process. If the first processing step is a compression process, the second processing step is a recovery process. If the first processing step is a recovery process, the second processing step is a compression process.

[0199] In some embodiments, the first processing step is a modulation or demodulation process. If the first processing step is a modulation process, the second processing step is a demodulation process. If the first processing step is a demodulation process, the second processing step is a modulation process.

[0200] In some embodiments, the first model and the second model can form a model pair, or in other words, the first model and the second model can form a bilateral model.

[0201] In some embodiments, the names of the first model and the second model are not limited, and may be, for example, "generation model", "recovery model", "encoder model", "decoder model", "compression model", "decompression model", "modulation model", "demodulation model", etc.

[0202] In some embodiments, the first model may be trained based on the third model.

[0203] In some embodiments, the second model may be trained based on the third model.

[0204] In some embodiments, the third model is a trained model, a deployed model, or a standardized model.

[0205] In some embodiments, the term "third model" is not limited, and may be, for example, "reference model," "standard model," "original model," "old model," etc.

[0206] In some embodiments, the third model can be used to perform the first processing procedure. Alternatively, the third model can be used to perform the second processing procedure. Or, the third model can be used to perform both the first and second processing procedures, in which case the third model comprises at least one model pair.

[0207] The following uses a two-sided model, with the encoding and decoding models as examples, to illustrate the model pairing process.

[0208] In this embodiment of the disclosure, the model pairing process includes: (1) a fourth node sending a third model and / or second information associated with the third model. (2) a third node training a first model based on the third model and / or the second information associated with the third model. (3) a second node configuring first information for the first model. (4) a first node determining the model pairing relationship based on the first information.

[0209] In some embodiments, when the fourth node and the third node are deployed on the same device, the device can directly train the first model based on the third model and / or the second information associated with the third model.

[0210] In some embodiments, when the fourth node and the third node are deployed on different devices, the fourth node sends the third model and / or the second information associated with the third model to the third node.

[0211] In one example, the fourth node is deployed on the network device, and the third node is deployed on both the terminal and the network device. The network device sends the third model and / or the second information associated with the third model to the terminal. Based on the third model and / or the second information associated with the third model, the terminal trains a first model, which is an encoding model. The network device then trains the first model again based on the third model and / or the second information associated with the third model, which is a decoding model. This can be understood as both the terminal and the network device retraining the first model based on the third model.

[0212] In some embodiments, when the third node and the second node are deployed on the same device, the device can directly configure the first information for the first model.

[0213] In some embodiments, when the third node and the second node are deployed on different devices, the third node may send a first message to the second node to request the second node to configure first information. The second node may then send the configured first information to the third node.

[0214] In some embodiments, when the third node and the first node are deployed on the same device, the device can directly determine the model pairing relationship based on the first information.

[0215] In some embodiments, when the third node and the first node are deployed on different devices, the third node may send a second message to the first node to trigger the first node to determine the model pairing relationship.

[0216] Figure 2A is a schematic diagram of a first interaction of a communication method provided according to an embodiment of the present disclosure. As shown in Figure 2A, the present disclosure relates to a communication method. Executed by a communication system 100, the communication method includes steps S2101 to S2110.

[0217] In this embodiment of the disclosure, the model pairing process and the model application process are illustrated by taking the deployment of the first node, the second node, the fourth node and the fifth node on the network device and the deployment of the third node on the terminal as an example.

[0218] In some embodiments, the first model obtained by retraining the terminal is an encoding model, and the second model paired with the first model is a decoding model.

[0219] In step S2101, the network device sends the third model and / or the second information.

[0220] In some embodiments, the terminal receives a third model.

[0221] In some embodiments, the third model is used to train the first model on the terminal.

[0222] In some embodiments, the third model includes encoding model A and / or decoding model A. Encoding model A and decoding model A are paired.

[0223] In some embodiments, the first model is an encoding model B. When the third model is an encoding model A, encoding model A can be used to train encoding model B. When the third model is a decoding model A, decoding model A can be used to train encoding model B. When the third model includes both encoding model A and decoding model A, encoding model A and decoding model A can be used to generate encoding model B.

[0224] In some embodiments, the second information is used by the terminal to train the first model.

[0225] In some embodiments, the second information includes at least one of the following: third information (denoted as model identifier A), fourth information (denoted as model structure information), fifth information (denoted as model parameter information), and sixth information (denoted as training data information).

[0226] In some embodiments, model identifier A is used to identify a third model, wherein model identifier A is configured by the network device for the third model.

[0227] In some embodiments, model structure information is used to determine the model structure A of the third model.

[0228] In some embodiments, model structure A is used to train a first model. In other words, the model structure B of the first model is the same as the model structure A of the third model.

[0229] In some embodiments, model parameter information is used to determine model parameters A of the third model.

[0230] In some embodiments, model parameters A are used to train a first model. In other words, the model parameters B of the first model are the same as the model parameters A of the third model.

[0231] In some embodiments, training data information is used to determine training data A for the third model. In some embodiments, training data A includes input data and / or output data of the third model.

[0232] In some embodiments, training data A is used to train a first model. In other words, the training data B of the first model is the same as the training data A of the third model.

[0233] In some embodiments, if the second information does not include training data information, the network device may send a configuration B associated with the type of training data to the terminal. The terminal may collect training data B based on the configuration B, and the training data B may be used to train a first model.

[0234] In some embodiments, configuration B includes at least one of the following: the number of transmit antenna ports, transmission bandwidth, model input data type, and quantization method of training data. Thus, the terminal or network device can determine the type of training data collected based on configuration B.

[0235] In some embodiments, if the second information does not include training data information, the network device may send a condition B for collecting training data to the terminal, and the terminal may collect training data B based on condition B. The training data B is used to train and obtain the first model.

[0236] In some embodiments, condition B includes at least one of the following: channel scenario, maximum rank supported by the terminal or network device, and mobile state information of the terminal.

[0237] In some embodiments, model identifier A includes at least one of the following: seventh indication information, eighth indication information, ninth indication information, tenth indication information, eleventh indication information, and twelfth indication information.

[0238] In some embodiments, the seventh indication information is used to indicate the third model. In other words, the seventh indication information is the identifier (ID) of the third model.

[0239] In one example, when the third model is encoding model A, the seventh indication information is the identifier of encoding model A (encoderA ID). When the third model is decoding model A, the seventh indication information is the identifier of decoding model A (decoderA ID). When the third model includes both encoding model A and decoding model A, the seventh indication information is the identifier of model pair A (pairedA ID).

[0240] In some embodiments, the eighth indication information is used to indicate the encoding model A and the decoding model A. In other words, the eighth indication information is an identifier for model A. It can be understood that the identifier for model A can be used to indicate the encoding model A, the decoding model A, or both the encoding model A and the decoding model A.

[0241] In some embodiments, the ninth indication information is used to indicate the training data A of the third model. In other words, the ninth indication information is the identifier (datasetA ID) of the training data A. It can be understood that the third model is indicated by the identifier of the training data of the third model.

[0242] In some embodiments, the tenth indication information is used to indicate the training session A of the third model. In other words, the tenth indication information is the identifier of training session A (training session A ID). It can be understood that the third model is indicated by the identifier of the training session of the third model.

[0243] In some embodiments, the eleventh indication information is used to indicate the configuration A associated with the type of the training data A of the third model. In other words, the eleventh indication information is an identifier of configuration A. It can be understood that the third model is indicated by the configuration identifier associated with the type of the training data of the third model.

[0244] In some embodiments, configuration A includes at least one of the following: the number of transmit antenna ports, transmission bandwidth, model input data type, and quantization method of training data. Thus, the terminal or network device can determine the type of training data collected based on configuration A. In some embodiments, configuration A is configured by the network device.

[0245] In some embodiments, the twelfth indication information is used to indicate condition A for collecting training data A of the third model. In other words, the twelfth indication information is an identifier for condition A. It can be understood that the third model is indicated by the identifier of the collection condition for the training data of the third model.

[0246] In some embodiments, condition A includes at least one of the following: channel scenario, maximum rank supported by the terminal or network device, and mobile state information of the terminal.

[0247] In some embodiments, the identifiers of configuration A and condition A can be the same (e.g., the identifiers of configuration A and condition A are both associatedA ID).

[0248] In some embodiments, network devices can transmit third models through radio resource control (RRC) signaling, medium access control control element (MAC-CE) signaling, downlink control information (DCI), etc.

[0249] In step S2102, the terminal trains the first model based on the third model and / or the second information.

[0250] In some embodiments, when the network device sends a third model, the terminal can train a first model based on the model structure A and model parameters A of the third model.

[0251] In some embodiments, when the second information sent by the network device includes a model identifier A, the terminal can obtain the third model based on the model identifier A that indicates the third model, and train the first model based on the third model.

[0252] In some embodiments, when the second information sent by the network device includes the model structure information, the terminal can train the first model based on the model structure A of the third model.

[0253] In some embodiments, when the second information sent by the network device includes model parameter information, the terminal can train the first model based on the model parameters A of the third model.

[0254] In some embodiments, when the second information sent by the network device includes training data information, the terminal can use the training data A of the third model to train the first model.

[0255] In some embodiments, if the second information sent by the network device does not include training data information, the terminal may collect training data for the first model based on the configuration B associated with the training data sent by the network device, and / or the condition B for collecting training data.

[0256] In some embodiments, the methods described above for training the first model can be used in combination without conflict, and will not be elaborated here.

[0257] In step S2103, the terminal sends the first message.

[0258] In some embodiments, the network device receives a first message.

[0259] In some embodiments, the first message is used to request first information (denoted as model identifier B) for the first model.

[0260] In some embodiments, model identifier B includes one of the following: first indication information, second indication information, third indication information, fourth indication information, fifth indication information, and sixth indication information.

[0261] In some embodiments, the first indication information is used to indicate a first model. In one example, the first indication information is the identifier (encoderB ID) of encoding model B.

[0262] In some embodiments, the second indication information is used to indicate a model pair B. The model pair B includes a first model (i.e., encoding model B) and a second model (i.e., decoding model B). This can be understood as indicating the first model through the identifier of the model pair B. In one example, the second indication information is the identifier of the model pair B (pairdeB ID).

[0263] In some embodiments, the third indication information is used to indicate the training data B of the first model. This can be understood as indicating the first model through the identifier of its training data. In one example, the third indication information is the identifier (datasetB ID) of the training data B.

[0264] In some embodiments, the fourth indication information is used to indicate the training session B of the first model. This can be understood as indicating the first model through the identifier of its training session. In one example, the fourth indication information is the identifier of training session B (training session B ID).

[0265] In some embodiments, the fifth indication information is used to indicate the configuration B associated with the type of the training data B of the first model. This can be understood as indicating the first model through the configuration identifier associated with the type of the training data of the first model. In one example, the fifth indication information is the identifier of configuration B (AssociatedB ID).

[0266] In some embodiments, the sixth indication information is used to indicate condition A for collecting training data A of the first model. This can be understood as indicating the first model through the identifier of the collection condition for the first model's training data. In one example, the sixth indication information is the identifier of condition B (AssociatedB ID).

[0267] In some embodiments, the first message is also used to notify the network device that the terminal has completed training of the first model.

[0268] In some embodiments, the name of the first message is not limited, and it may be, for example, "training complete message", "ID request message", etc.

[0269] In some embodiments, step S2103 can be omitted. In this case, the network device can send the third model and model identifier B together to the terminal during the execution of step S2101, or send the second information and model identifier B together to the terminal, or send the third model, the second information, and model identifier B together to the terminal. This can save signaling interaction between the terminal and the network device and improve the efficiency of model pairing. However, at the same time, the network device may not be able to know temporarily whether the terminal has retrained a new model.

[0270] In step S2104, the network device sends model identifier B.

[0271] In some embodiments, the terminal receives model identifier B.

[0272] In some embodiments, steps S2104 and S2101 can be executed simultaneously. In some embodiments, step S2104 can also be executed before step S2102. That is, the network device can configure the model identifier B in advance before the terminal trains the first model, thereby saving signaling interaction between the terminal and the network device and improving the efficiency of configuring the identifier and matching the model.

[0273] In some embodiments, steps S2103 and S2104 can both be omitted. In this case, the terminal can determine model identifier A, which indicates the third model, as model identifier B, that is, model identifier B is the same as model identifier A. In other words, model identifier A indicates both the third model and the first model. In this way, the number of model identifiers can be reduced, the complexity of maintaining model identifiers in the system can be reduced, and the pairing efficiency can be improved.

[0274] In some embodiments, the terminal or network device may calculate and determine the model identifier B based on an identifier specified in the protocol or configured in the network and a first value. In one example, the identifier specified in the protocol or configured in the network may be one or more of encoder ID, paired ID, dataset ID, training session ID, and associated ID, and the first value is X. Taking the identifier specified in the protocol or configured in the network as encoder ID as an example, the model identifier B is encoder ID + X.

[0275] In some embodiments, the identifier specified in the protocol or configured in the network may be model identifier A.

[0276] In step S2105, the terminal sends a second message.

[0277] In some embodiments, the network device receives a second message.

[0278] In some embodiments, the second message is used to trigger the model pairing process.

[0279] In some embodiments, the second message carries a model identifier B, in which case the second message is used to trigger pairing of the first model indicated by the model identifier B.

[0280] In some embodiments, the name of the second message is not limited, and it may be, for example, "pairing request message", "pairing trigger message", "pairing start message", etc.

[0281] In some embodiments, steps S2102 to S2105 may be omitted, in which case CWN determines the first frequency point based on network indication.

[0282] In step S2106, the network device determines the model pairing relationship based on model identifier B.

[0283] In some embodiments, the encoding model B indicated by model identifier B is trained based on a third model, which may include encoding model A and / or decoding model A. Therefore, the network device determines that the decoding model B paired with the encoding model B is decoding model A, establishing a pairing relationship between the model identifier B of the encoding model B and the model identifier A of the decoding model A. In one example, the model pairing relationship includes: encoderB ID - decoderA ID. In another example, the model pairing relationship also includes: encoderA ID - decoderA ID.

[0284] In some embodiments, when the encoderA ID, pairedA ID, datasetA ID, and trainingA session ID of encoding model A are the same as the encoderB ID, pairedB ID, datasetB ID, and trainingB session ID of encoding model B, the network device can use the associatedB ID to determine the model pairing relationship since the associated IDs of encoding model A and encoding model B are different.

[0285] In step S2107, the terminal uses the first model to process data and obtain the first data.

[0286] In some embodiments, when multiple models are deployed on the terminal, the terminal can select one model for data processing. This model can be a standard model or reference model deployed by the network device, or a new model obtained by retraining based on the standard model or reference model.

[0287] In step S2108, the terminal sends model identifier B and first data.

[0288] In some embodiments, the network device receives model identifier B and first data.

[0289] In some embodiments, model identifier B is used by the network device to determine a second model paired with the first model based on the model pairing relationship.

[0290] In step S2109, the network device determines the second model paired with the first model based on the model identifier B and the model pairing relationship.

[0291] In some embodiments, model pairing relationships are used to indicate at least one model pair, each model pair including an encoded model and a decoded model. The network device previously determines model pairing relationships, and based on model identifier B, searches for a second model from the model pairing relationships.

[0292] In step S2110, the network device uses the second model to process the first data.

[0293] In this embodiment of the disclosure, steps S2101 to S2106 are model pairing processes, and steps S2107 to S2110 are model application processes. The two processes can be executed separately.

[0294] In some embodiments, the first node, the second node, the fourth node, and the fifth node may also be deployed on the terminal, and the third node may also be deployed on the network device.

[0295] In some embodiments, the first model obtained by retraining the network device is a decoding model, and the second model paired with the first model is an encoding model.

[0296] In some embodiments, the network device in steps S2101 to S2106 can be replaced with a terminal, and the terminal in steps S2101 to S2106 can be replaced with a network device.

[0297] In some embodiments, the terminal may send a third model, second information, model identifier B, first message or second message to the network device via RRC signaling, uplink control information (DCI), physical uplink control channel (PUCCH) and physical uplink shared channel (PUSCH).

[0298] In some embodiments, when the first node is deployed on the terminal (i.e., the terminal determines the model pairing relationship), the network device cannot determine the second model paired with the first model. Therefore, during step S2108, the terminal can send the first data and the seventh information indicating the second model to the network device so that the network device can process the first data using the second model. Alternatively, the terminal can send the model identifier B, the first data, and the eighth information indicating the model pairing relationship to the network device so that the network device can determine the second model to use based on the model pairing relationship.

[0299] In some embodiments, the terminal may send the model identifier B, the first data, and the eighth information to the network device during the execution of step S2108. Alternatively, the terminal may send the eighth information to the network device after step S2106 and before step S2107.

[0300] In some embodiments, the seventh information indicating the second model is determined by the terminal from the model pairing relationships it determines.

[0301] The communication method involved in the embodiments of this disclosure may include at least one of steps S2101 to S2110. For example, step S2101 may be implemented as a standalone embodiment. For example, step S2102 may be implemented as a standalone embodiment. For example, step S2103 may be implemented as a standalone embodiment. For example, step S2104 may be implemented as a standalone embodiment. For example, step S2105 may be implemented as a standalone embodiment. For example, step S2106 may be implemented as a standalone embodiment. For example, step S2107 may be implemented as a standalone embodiment. For example, step S2108 may be implemented as a standalone embodiment. For example, step S2109 may be implemented as a standalone embodiment. For example, step S2110 may be implemented as a standalone embodiment. For example, steps S2101 and S2104 may be combined as a standalone embodiment. For example, steps S2104, S2105, and S2106 may be combined as a standalone embodiment. For example, steps S2108, S2109 and S2110 can be combined as independent embodiments.

[0302] Figure 2B is a second interactive schematic diagram of the communication method provided according to an embodiment of the present disclosure. As shown in Figure 2B, the present disclosure relates to a communication method. Executed by a communication system 100, the communication method includes steps S2201 to S2209.

[0303] In this embodiment of the disclosure, the model pairing process and the model application process are illustrated by taking the deployment of the first node, the fourth node and the fifth node on the network device and the deployment of the second node and the third node on the terminal as an example.

[0304] In some embodiments, the first model obtained by retraining the terminal is an encoding model, and the second model paired with the first model is a decoding model.

[0305] In step S2201, the network device sends the third model and / or the second information.

[0306] Other optional implementations of step S2201 can be found in the optional implementations of step S2102 in Figure 2A and other related parts in the embodiments involved in Figure 2A, which will not be repeated here.

[0307] In step S2202, the terminal trains the first model based on the third model and / or the second information.

[0308] Other optional implementations of step S2202 can be found in the optional implementations of step S2102 in Figure 2A and other related parts in the embodiments involved in Figure 2A, which will not be repeated here.

[0309] In step S2203, the terminal configures model identifier B for the first model.

[0310] In some embodiments, the terminal may determine model identifier A, which indicates the third model, as model identifier B, meaning that model identifier B is the same as model identifier A. In other words, model identifier A indicates both the third model and the first model. This reduces the number of model identifiers, lowers the complexity of maintaining model identifiers in the system, and improves pairing efficiency.

[0311] In some embodiments, the terminal may also calculate and determine the model identifier B based on an identifier specified in the protocol or configured in the network and a first value. In one example, the type of the identifier specified in the protocol or configured in the network may be one or more of encoder ID, paired ID, dataset ID, training session ID, and associated ID, and the first value is X. Taking the identifier specified in the protocol or configured in the network as encoder ID as an example, the model identifier B is encoder ID + X.

[0312] In some embodiments, the identifier specified in the protocol or configured in the network may be model identifier A.

[0313] In step S2204, the terminal sends a second message.

[0314] Other optional implementations of step S2204 can be found in the optional implementations of step S2105 in Figure 2A and other related parts in the embodiments involved in Figure 2A, which will not be repeated here.

[0315] In step S2205, the network device determines the model pairing relationship based on model identifier B.

[0316] Other optional implementations of step S2205 can be found in the optional implementations of step S2106 in Figure 2A and other related parts in the embodiments involved in Figure 2A, which will not be repeated here.

[0317] In step S2206, the terminal uses the first model to process data and obtain the first data.

[0318] Other optional implementations of step S2206 can be found in the optional implementations of step S2107 in Figure 2A and other related parts in the embodiments involved in Figure 2A, which will not be repeated here.

[0319] In step S2207, the terminal sends model identifier B and first data.

[0320] Other optional implementations of step S2207 can be found in the optional implementations of step S2108 in Figure 2A and other related parts in the embodiments involved in Figure 2A, which will not be repeated here.

[0321] In step S2208, the network device determines the second model paired with the first model based on the model identifier B and the model pairing relationship.

[0322] Other optional implementations of step S2208 can be found in the optional implementations of step S2109 in Figure 2A and other related parts in the embodiments involved in Figure 2A, which will not be repeated here.

[0323] In step S2209, the network device uses the second model to process the first data.

[0324] Other optional implementations of step S2209 can be found in the optional implementations of step S2110 in Figure 2A and other related parts in the embodiments involved in Figure 2A, which will not be repeated here.

[0325] In this embodiment of the disclosure, steps S2201 to S2205 are model pairing processes, and steps S2206 to S2209 are model application processes. The two processes can be executed separately.

[0326] In some embodiments, the first node, the fourth node, and the fifth node may also be deployed on the terminal, and the second node and the third node may also be deployed on the network device.

[0327] In some embodiments, the first model obtained by retraining the network device is a decoding model, and the second model paired with the first model is an encoding model.

[0328] In some embodiments, the execution entity network device in steps S2201 to S2205 can be replaced with a terminal, and the execution entity terminal in steps S2201 to S2205 can be replaced with a network device.

[0329] In some embodiments, when the first node is deployed on the terminal (i.e., the terminal determines the model pairing relationship), the network device cannot determine the second model paired with the first model. Therefore, during the execution of step S2207, the terminal can send the first data and the seventh information indicating the second model to the network device so that the network device can process the first data using the second model. Alternatively, the terminal can send the model identifier B, the first data, and the eighth information indicating the model pairing relationship to the network device so that the network device can determine the second model to use based on the model pairing relationship.

[0330] In some embodiments, the terminal may send the model identifier B, the first data, and the eighth information to the network device during the execution of step S2207. Alternatively, the terminal may send the eighth information to the network device after step S2205 and before step S2206.

[0331] In some embodiments, the seventh information indicating the second model is determined by the terminal from the model pairing relationships it determines.

[0332] The communication method involved in the embodiments of this disclosure may include at least one of steps S2201 to S2209. For example, step S2201 may be implemented as a standalone embodiment. For example, step S2202 may be implemented as a standalone embodiment. For example, step S2203 may be implemented as a standalone embodiment. For example, step S2204 may be implemented as a standalone embodiment. For example, step S2205 may be implemented as a standalone embodiment. For example, step S2206 may be implemented as a standalone embodiment. For example, step S2207 may be implemented as a standalone embodiment. For example, step S2208 may be implemented as a standalone embodiment. For example, step S2209 may be implemented as a standalone embodiment. For example, steps S2203, S2204, and S2205 may be combined as a standalone embodiment. For example, steps S2207, S2208, and S2209 may be combined as a standalone embodiment. However, this is not the limitation.

[0333] Figure 2C is a third interactive schematic diagram of the communication method provided according to an embodiment of the present disclosure. As shown in Figure 2C, the present disclosure relates to a communication method. Executed by a communication system 100, the communication method includes steps S2301 to S2308.

[0334] In this embodiment of the disclosure, the model pairing process and the model application process are illustrated by taking the deployment of the second and fourth nodes on network devices and the deployment of the first, third, and fifth nodes on terminals as examples.

[0335] In some embodiments, the first model obtained by retraining the terminal is an encoding model, and the second model paired with the first model is a decoding model.

[0336] In step S2301, the network device sends the third model and / or the second information.

[0337] Other optional implementations of step S2301 can be found in the optional implementations of step S2101 in Figure 2A and other related parts in the embodiments involved in Figure 2A, which will not be repeated here.

[0338] In step S2302, the terminal trains the first model based on the third model and / or the second information.

[0339] Other optional implementations of step S2302 can be found in the optional implementations of step S2102 in Figure 2A and other related parts in the embodiments involved in Figure 2A, which will not be repeated here.

[0340] In step S2303, the terminal sends the first message.

[0341] Other optional implementations of step S2303 can be found in the optional implementations of step S2103 in Figure 2A and other related parts in the embodiments involved in Figure 2A, which will not be repeated here.

[0342] In step S2304, the network device sends model identifier B.

[0343] Other optional implementations of step S2304 can be found in the optional implementations of step S2104 in Figure 2A and other related parts in the embodiments involved in Figure 2A, which will not be repeated here.

[0344] In step S2305, the terminal determines the model pairing relationship based on model identifier B.

[0345] In some embodiments, the encoding model B indicated by model identifier B is trained based on a third model, which may include encoding model A and / or decoding model A. Therefore, the terminal determines that the decoding model B paired with the encoding model B is decoding model A, and establishes a pairing relationship between the model identifier B of encoding model B and the model identifier A of decoding model A. In one example, the model pairing relationship includes: encoderB ID - decoderA ID. In another example, the model pairing relationship also includes: encoderA ID - decoderA ID.

[0346] In some embodiments, when the encoderA ID, pairedA ID, datasetA ID, and trainingA session ID of encoding model A are the same as the encoderB ID, pairedB ID, datasetB ID, and trainingB session ID of encoding model B, the network device can use the associatedB ID to determine the model pairing relationship since the associated IDs of encoding model A and encoding model B are different.

[0347] In step S2306, the terminal uses the first model to process data and obtain the first data.

[0348] Other optional implementations of step S2306 can be found in the optional implementations of step S2107 in Figure 2A and other related parts in the embodiments involved in Figure 2A, which will not be repeated here.

[0349] In step S2307, the terminal sends the first data and the seventh information.

[0350] In some embodiments, the seventh information is used to indicate the second model paired with the first model.

[0351] In some embodiments, the seventh information indicating the second model is determined by the terminal from the model pairing relationships it determines.

[0352] In some embodiments, the terminal may send model identifier B, first data, and eighth information. The eighth information is used to indicate the model pairing relationship.

[0353] Other optional implementations of step S2307 can be found in the optional implementations of step S2108 in Figure 2A and other related parts in the embodiments involved in Figure 2A, which will not be repeated here.

[0354] In step S2308, the network device uses the second model to process the first data.

[0355] Other optional implementations of step S2308 can be found in the optional implementations of step S2109 in Figure 2A and other related parts in the embodiments involved in Figure 2A, which will not be repeated here.

[0356] In this embodiment of the disclosure, steps S2301 to S2305 are model pairing processes, and steps S2306 to S2308 are model application processes. The two processes can be executed separately.

[0357] In some embodiments, the second and fourth nodes may also be deployed on the terminal, and the first, third, and fifth nodes may also be deployed on the network device. In this case, the first model retrained by the network device is a decoding model, and the second model paired with the first model is an encoding model.

[0358] In some embodiments, the execution entity network device in steps S2301, S2302 and S2305 can be replaced with a terminal, and the execution entity terminal in steps S2301, S2302 and S2305 can be replaced with a network device.

[0359] In some embodiments, when the functionality of the first node is deployed on the network device (i.e., the network device determines the model pairing relationship), the terminal only needs to send the model identifier B and the first data during the execution of step S2307. The network device can determine the second model to be used based on the previously determined model pairing relationship.

[0360] The communication method involved in the embodiments of this disclosure may include at least one of steps S2301 to S2308. For example, step S2301 may be implemented as a standalone embodiment. For example, step S2302 may be implemented as a standalone embodiment. For example, step S2303 may be implemented as a standalone embodiment. For example, step S2304 may be implemented as a standalone embodiment. For example, step S2305 may be implemented as a standalone embodiment. For example, step S2306 may be implemented as a standalone embodiment. For example, step S2307 may be implemented as a standalone embodiment. For example, step S2308 may be implemented as a standalone embodiment. For example, steps S2303, S2304, and S2305 may be combined as a standalone embodiment. For example, steps S2307 and S2308 may be combined as a standalone embodiment. However, this is not the limitation.

[0361] In some embodiments, steps S2301 and S2304 can be performed simultaneously.

[0362] In some embodiments, steps S2303 and S2304 may be omitted.

[0363] Figure 2D is a fourth interactive schematic diagram of the communication method provided according to an embodiment of the present disclosure. As shown in Figure 2D, the present disclosure relates to a communication method. Executed by a communication system 100, the communication method includes steps S2401 to S2408.

[0364] In this embodiment of the disclosure, the model pairing process and the model application process are illustrated by taking the fourth node deployed on a network device and the first node, second node, third node and fifth node deployed on a terminal as an example.

[0365] In some embodiments, the first model obtained by retraining the terminal is an encoding model, and the second model paired with the first model is a decoding model.

[0366] In step S2401, the network device sends the third model and / or the second information.

[0367] Other optional implementations of step S2401 can be found in the optional implementations of step S2101 in Figure 2A and other related parts in the embodiments involved in Figure 2A, which will not be repeated here.

[0368] In step S2402, the terminal trains the first model based on the third model and / or the second information.

[0369] Other optional implementations of step S2402 can be found in the optional implementations of step S2102 in Figure 2A and other related parts in the embodiments involved in Figure 2A, which will not be repeated here.

[0370] In step S2403, the terminal configures model identifier B for the first model.

[0371] In some embodiments, the terminal may determine model identifier A, which indicates the third model, as model identifier B, meaning that model identifier B is the same as model identifier A. In other words, model identifier A indicates both the third model and the first model. This reduces the number of model identifiers, lowers the complexity of maintaining model identifiers in the system, and improves pairing efficiency.

[0372] In some embodiments, the terminal may also calculate and determine the model identifier B based on an identifier specified in the protocol or configured in the network and a first value. In one example, the type of the identifier specified in the protocol or configured in the network may be one or more of encoder ID, paired ID, dataset ID, training session ID, and associated ID, and the first value is X. Taking the identifier specified in the protocol or configured in the network as encoder ID as an example, the model identifier B is encoder ID + X.

[0373] In some embodiments, the identifier specified in the protocol or configured in the network may be model identifier A.

[0374] In step S2404, the terminal determines the model pairing relationship based on model identifier B.

[0375] In some embodiments, the encoding model B indicated by model identifier B is trained based on a third model, which may include encoding model A and / or decoding model A. Therefore, the terminal determines that the decoding model B paired with the encoding model B is decoding model A, and establishes a pairing relationship between the model identifier B of encoding model B and the model identifier A of decoding model A. In one example, the model pairing relationship includes: encoderB ID - decoderA ID. In another example, the model pairing relationship also includes: encoderA ID - decoderA ID.

[0376] In some embodiments, when the encoderA ID, pairedA ID, datasetA ID, and trainingA session ID of encoding model A are the same as the encoderB ID, pairedB ID, datasetB ID, and trainingB session ID of encoding model B, the network device can use the associatedB ID to determine the model pairing relationship since the associated IDs of encoding model A and encoding model B are different.

[0377] In step S2405, the terminal uses the first model to process data and obtain the first data.

[0378] Other optional implementations of step S2405 can be found in the optional implementations of step S2107 in Figure 2A and other related parts in the embodiments involved in Figure 2A, which will not be repeated here.

[0379] In step S2406, the terminal sends model identifier B, first data, and eighth information.

[0380] In some embodiments, the terminal sends first data and seventh information.

[0381] Other optional implementations of step S2406 can be found in the optional implementations of step S2108 in Figure 2A and other related parts in the embodiments involved in Figure 2A, which will not be repeated here.

[0382] In step S2407, the network device determines the second model paired with the first model based on the model identifier B and the model pairing relationship.

[0383] Other optional implementations of step S2407 can be found in the optional implementations of step S2109 in Figure 2A and other related parts in the embodiments involved in Figure 2A, which will not be repeated here.

[0384] In step S2408, the network device uses the second model to process the first data.

[0385] Other optional implementations of step S2408 can be found in the optional implementations of step S2109 in Figure 2A and other related parts in the embodiments involved in Figure 2A, which will not be repeated here.

[0386] In this embodiment of the disclosure, steps S2401 to S2404 are model pairing processes, and steps S2405 to S2408 are model application processes. The two processes can be executed separately.

[0387] In some embodiments, the fourth node may also be deployed on a terminal, and the first, second, third, and fifth nodes may also be deployed on a network device. In this case, the first model retrained by the network device is a decoding model, and the second model paired with the first model is an encoding model.

[0388] In some embodiments, the execution entity network device in steps S2401 to S2404 can be replaced with a terminal, and the execution entity terminal in steps S2401 to S2404 can be replaced with a network device.

[0389] In some embodiments, when the function of the first node is deployed on the network device (i.e. the network device determines the model pairing relationship), the terminal only needs to send the model identifier B and the first data during the execution of step S2406. The network device can determine the second model to be used based on the previously determined model pairing relationship.

[0390] The communication method involved in the embodiments of this disclosure may include at least one of steps S2401 to S2408. For example, step S2401 may be implemented as a standalone embodiment. For example, step S2402 may be implemented as a standalone embodiment. For example, step S2403 may be implemented as a standalone embodiment. For example, step S2404 may be implemented as a standalone embodiment. For example, step S2405 may be implemented as a standalone embodiment. For example, step S2406 may be implemented as a standalone embodiment. For example, step S2407 may be implemented as a standalone embodiment. For example, step S2408 may be implemented as a standalone embodiment.

[0391] For example, steps S2403 and S2404 can be combined as independent embodiments. For example, steps S2406, S2407, and S2408 can be combined as independent embodiments.

[0392] In some embodiments, the terms "bilateral model", "model pair", and "model group" can be used interchangeably.

[0393] In some embodiments, terms such as “pairing,” “matching,” and “association” can be used interchangeably.

[0394] In some embodiments, the terms "model application", "model reasoning", and "model usage" can be used interchangeably.

[0395] In some embodiments, the names of information, etc., are not limited to the names described in the embodiments. Terms such as "information", "message", "signal", "signaling", "report", "configuration", "indication", "instruction", "command", "channel", "parameter", "domain", "field", "symbol", "symbol", "codebook", "codeword", "codepoint", "bit", "data", "program", and "chip" can be used interchangeably.

[0396] In some embodiments, the terms “carrying,” “including,” “containing,” and “encapsulating” can be used interchangeably.

[0397] In some embodiments, the terms “radio”, “wireless”, “radioaccessnetwork (RAN)”, “accessnetwork (AN)”, and “RAN-based” can be used interchangeably.

[0398] In some embodiments, “get,” “obtain,” “receive,” “transmit,” “bidirectional transmission,” and “send and / or receive” can be used interchangeably and can be interpreted as receiving from other entities, obtaining from protocols, obtaining from higher layers, obtaining through self-processing, or autonomous implementation, among other meanings.

[0399] In some embodiments, terms such as “send,” “transmit,” “report,” “transmit,” “request,” “bidirectional transmission,” “send and / or receive” can be used interchangeably.

[0400] In some embodiments, the terms “issue,” “return,” “feedback,” “response,” and “acknowledgement” can be used interchangeably.

[0401] In some embodiments, terms such as "certain," "preset," "default," "set," "indicated," "a certain," "any," and "first" can be used interchangeably. "Certain A," "preset A," "default A," "set A," "indicated A," "a certain A," "any A," and "first A" can be interpreted as A pre-defined in a protocol or the like, or as A obtained through setting, configuration, or instruction, or as specific A, a certain A, any A, or first A, but are not limited thereto.

[0402] In some embodiments, the determination or judgment can be made by a value represented by 1 bit (0 or 1), or by a true or false value (boolean), or by a comparison of numerical values ​​(e.g., a comparison with a predetermined value), but is not limited thereto.

[0403] Figure 3A is a schematic flowchart illustrating a first type of communication method executed by a terminal according to an embodiment of the present disclosure. As shown in Figure 3A, the present disclosure relates to a communication method executed by a terminal. The communication method includes steps S3101 to S3107.

[0404] In step S3101, the third model and / or the second information are received.

[0405] The optional implementation of step S3101 can be found in the optional implementation of step S2101 in Figure 2A and other related parts in the embodiments involved in Figure 2A, which will not be repeated here.

[0406] In step S3102, the first model is trained based on the third model and / or the second information.

[0407] The optional implementation of step S3102 can be found in the optional implementation of step S2102 in Figure 2A and other related parts in the embodiments involved in Figure 2A, which will not be repeated here.

[0408] In step S3103, the first message is sent.

[0409] The optional implementation of step S3103 can be found in the optional implementation of step S2103 in Figure 2A and other related parts in the embodiments involved in Figure 2A, which will not be repeated here.

[0410] In step S3104, model identifier B is received.

[0411] The optional implementation of step S3104 can be found in the optional implementation of step S2104 in Figure 2A and other related parts in the embodiments involved in Figure 2A, which will not be repeated here.

[0412] In step S3105, a second message is sent.

[0413] The optional implementation of step S3105 can be found in the optional implementation of step S2105 in Figure 2A and other related parts in the embodiments involved in Figure 2A, which will not be repeated here.

[0414] In step S3106, the first model is used to process the data to obtain the first data.

[0415] The optional implementation of step S3106 can be found in the optional implementation of step S2107 in Figure 2A and other related parts in the embodiments involved in Figure 2A, which will not be repeated here.

[0416] In step S3107, model identifier B and first data are sent.

[0417] The optional implementation of step S3107 can be found in the optional implementation of step S2108 in Figure 2A and other related parts in the embodiments involved in Figure 2A, which will not be repeated here.

[0418] Figure 3B is a schematic diagram of the second final process of a terminal executing a communication method according to an embodiment of the present disclosure. As shown in Figure 3B, the embodiments of the present disclosure relate to a communication method executed by a terminal. The above-mentioned communication method includes steps S3201 to S3206.

[0419] In step S3201, the third model is received.

[0420] The optional implementation of step S3201 can be found in the optional implementation of step S2201 in Figure 2B and other related parts in the embodiments involved in Figure 2B, which will not be repeated here.

[0421] In step S3202, the first model is obtained by training based on the third model.

[0422] The optional implementation of step S3202 can be found in the optional implementation of step S2202 in Figure 2B and other related parts in the embodiments involved in Figure 2B, which will not be repeated here.

[0423] In step S3203, a model identifier B is configured for the first model.

[0424] The optional implementation of step S3203 can be found in the optional implementation of step S2203 in Figure 2B and other related parts in the embodiments involved in Figure 2B, which will not be repeated here.

[0425] In step S3204, a second message is sent.

[0426] The optional implementation of step S3204 can be found in the optional implementation of step S2204 in Figure 2B and other related parts in the embodiments involved in Figure 2B, which will not be repeated here.

[0427] In step S3205, the first model is used for data processing to obtain the first data.

[0428] The optional implementation of step S3205 can be found in the optional implementation of step S2206 in Figure 2B and other related parts in the embodiments involved in Figure 2B, which will not be repeated here.

[0429] In step S3206, model identifier B and first data are sent.

[0430] The optional implementation of step S3206 can be found in the optional implementation of step S2207 in Figure 2B and other related parts in the embodiments involved in Figure 2B, which will not be repeated here.

[0431] Figure 3C is a schematic flowchart illustrating a third type of communication method executed by a terminal according to an embodiment of the present disclosure. As shown in Figure 3C, the embodiments of the present disclosure relate to a communication method executed by a terminal. The communication method includes steps S3301 to S3307.

[0432] In step S3301, the third model and / or the second information are received.

[0433] The optional implementation of step S3301 can be found in the optional implementation of step S2301 in Figure 2C and other related parts in the embodiments involved in Figure 2C, which will not be repeated here.

[0434] In step S3302, the first model is trained based on the third model and / or the second information.

[0435] The optional implementation of step S3302 can be found in the optional implementation of step S2302 in Figure 2C and other related parts in the embodiment involved in Figure 2C, which will not be repeated here.

[0436] In step S3303, the first message is sent.

[0437] For optional implementations of step S3303, please refer to the optional implementations of step S2303 in Figure 2C and other related parts in the embodiments involved in Figure 2C.

[0438] In step S3304, model identifier B is received.

[0439] The optional implementation of step S3304 can be found in the optional implementation of step S2304 in Figure 2C and other related parts in the embodiments involved in Figure 2C, which will not be repeated here.

[0440] In step S3305, the model pairing relationship is determined based on model identifier B.

[0441] The optional implementation of step S3305 can be found in the optional implementation of step S2305 in Figure 2C and other related parts in the embodiments involved in Figure 2C, which will not be repeated here.

[0442] In step S3306, the first model is used to process the data to obtain the first data.

[0443] The optional implementation of step S3306 can be found in the optional implementation of step S2306 in Figure 2C and other related parts in the embodiments involved in Figure 2C, which will not be repeated here.

[0444] In step S3307, the first data and the seventh information are sent.

[0445] The optional implementation of step S3307 can be found in the optional implementation of step S2307 in Figure 2C and other related parts in the embodiments involved in Figure 2C, which will not be repeated here.

[0446] Figure 3D is a schematic flowchart illustrating a fourth type of communication method executed by a terminal according to an embodiment of the present disclosure. As shown in Figure 3D, the present disclosure relates to a communication method executed by a terminal. The communication method includes steps S3401 to S3406.

[0447] In step S3401, the third model and / or the second information are received.

[0448] The optional implementation of step S3401 can be found in the optional implementation of step S2401 in Figure 2D and other related parts in the embodiments involved in Figure 2D, which will not be repeated here.

[0449] In step S3402, the first model is trained based on the third model and / or the second information.

[0450] The optional implementation of step S3402 can be found in the optional implementation of step S2402 in Figure 2D and other related parts in the embodiments involved in Figure 2D, which will not be repeated here.

[0451] In step S3403, a model identifier B is configured for the first model.

[0452] The optional implementation of step S3403 can be found in the optional implementation of step S2403 in Figure 2D and other related parts in the embodiments involved in Figure 2D, which will not be repeated here.

[0453] In step S3404, the model pairing relationship is determined based on model identifier B.

[0454] The optional implementation of step S3404 can be found in the optional implementation of step S2404 in Figure 2D and other related parts in the embodiments involved in Figure 2D, which will not be repeated here.

[0455] In step S3405, the first model is used for data processing to obtain the first data.

[0456] The optional implementation of step S3405 can be found in the optional implementation of step S2405 in Figure 2D and other related parts in the embodiments involved in Figure 2D, which will not be repeated here.

[0457] In step S3406, model identifier B, first data, and eighth information are sent.

[0458] The optional implementation of step S3406 can be found in the optional implementation of step S2406 in Figure 2D and other related parts in the embodiments involved in Figure 2D, which will not be repeated here.

[0459] Figure 3E is a schematic flowchart illustrating a first type of communication method performed by a network device according to an embodiment of the present disclosure. As shown in Figure 3E, the present disclosure relates to a communication method performed by a network device. The communication method includes steps S3501 to S3508.

[0460] In step S3501, the third model and / or the second information are sent.

[0461] The optional implementation of step S3501 can be found in the optional implementation of step S2101 in Figure 2A and other related parts in the embodiments involved in Figure 2A, which will not be repeated here.

[0462] In step S3502, the first message is received.

[0463] The optional implementation of step S3502 can be found in the optional implementation of step S2103 in Figure 2A and other related parts in the embodiments involved in Figure 2A, which will not be repeated here.

[0464] In step S3503, model identifier B is sent.

[0465] The optional implementation of step S3503 can be found in the optional implementation of step S2104 in Figure 2A and other related parts in the embodiments involved in Figure 2A, which will not be repeated here.

[0466] In step S3504, a second message is received.

[0467] The optional implementation of step S3504 can be found in the optional implementation of step S2105 in Figure 2A and other related parts in the embodiments involved in Figure 2A, which will not be repeated here.

[0468] In step S3505, the model pairing relationship is determined based on model identifier B.

[0469] The optional implementation of step S3505 can be found in the optional implementation of step S2106 in Figure 2A and other related parts in the embodiments involved in Figure 2A, which will not be repeated here.

[0470] In step S3506, model identifier B and first data are received.

[0471] The optional implementation of step S3506 can be found in the optional implementation of step S2108 in Figure 2A and other related parts in the embodiments involved in Figure 2A, which will not be repeated here.

[0472] In step S3507, a second model paired with the first model is determined based on the model identifier B and the model pairing relationship.

[0473] The optional implementation of step S3507 can be found in the optional implementation of step S2109 in Figure 2A and other related parts in the embodiments involved in Figure 2A, which will not be repeated here.

[0474] In step S3508, the first data is processed using the second model.

[0475] The optional implementation of step S3508 can be found in the optional implementation of step S2110 in Figure 2A and other related parts in the embodiments involved in Figure 2A, which will not be repeated here.

[0476] Figure 3F is a schematic flowchart illustrating a second method for a network device to perform a communication method according to an embodiment of the present disclosure. As shown in Figure 3F, the present disclosure relates to a communication method performed by a network device. The communication method includes steps S3601 to S3606.

[0477] In step S3601, the third model and / or the second information are sent.

[0478] The optional implementation of step S3601 can be found in the optional implementation of step S2201 in Figure 2B and other related parts in the embodiments involved in Figure 2B, which will not be repeated here.

[0479] In step S3602, a second message is received.

[0480] The optional implementation of step S3602 can be found in the optional implementation of step S2204 in Figure 2B and other related parts in the embodiments involved in Figure 2B, which will not be repeated here.

[0481] In step S3603, the model pairing relationship is determined based on model identifier B.

[0482] The optional implementation of step S3603 can be found in the optional implementation of step S2205 in Figure 2B and other related parts in the embodiments involved in Figure 2B, which will not be repeated here.

[0483] In step S3604, model identifier B and first data are received.

[0484] The optional implementation of step S3604 can be found in the optional implementation of step S2207 in Figure 2B and other related parts in the embodiments involved in Figure 2B, which will not be repeated here.

[0485] In step S3605, based on model identifier B and model pairing relationship, a second model paired with the first model is determined.

[0486] The optional implementation of step S3605 can be found in the optional implementation of step S2208 in Figure 2B and other related parts in the embodiments involved in Figure 2B, which will not be repeated here.

[0487] In step S3606, the second model is used to process the first data.

[0488] The optional implementation of step S3606 can be found in the optional implementation of step S2209 in Figure 2B, and other related parts in the embodiments involved in Figure 2B, which will not be repeated here.

[0489] Figure 3G is a schematic flowchart illustrating a third type of communication method performed by a network device according to an embodiment of the present disclosure. As shown in Figure 3G, the embodiments of the present disclosure relate to a communication method performed by a network device. The communication method includes steps S3701 to S3705.

[0490] In step S3701, the third model and / or the second information are sent.

[0491] The optional implementation of step S3701 can be found in the optional implementation of step S2301 in Figure 2C and other related parts in the embodiments involved in Figure 2C, which will not be repeated here.

[0492] In step S3702, the first message is received.

[0493] The optional implementation of step S3702 can be found in the optional implementation of step S2303 in Figure 2C and other related parts in the embodiments involved in Figure 2C, which will not be repeated here.

[0494] In step S3703, model identifier B is sent.

[0495] For optional implementations of step S3703, please refer to the optional implementations of step S2304 in Figure 2C and other related parts in the embodiments involved in Figure 2C.

[0496] In step S3704, the first data and the seventh information are received.

[0497] The optional implementation of step S3704 can be found in the optional implementation of step S2307 in Figure 2C and other related parts in the embodiments involved in Figure 2C, which will not be repeated here.

[0498] In step S3705, the first data is processed using the second model.

[0499] The optional implementation of step S3705 can be found in the optional implementation of step S2308 in Figure 2C and other related parts in the embodiments involved in Figure 2C, which will not be repeated here.

[0500] Figure 3H is a schematic flowchart illustrating a fourth type of communication method performed by a network device according to an embodiment of the present disclosure. As shown in Figure 3H, the embodiments of the present disclosure relate to a communication method performed by a network device. The communication method includes steps S3801 to S3804.

[0501] In step S3801, the third model and / or the second information are sent.

[0502] The optional implementation of step S3801 can be found in the optional implementation of step S2401 in Figure 2D and other related parts in the embodiments involved in Figure 2D, which will not be repeated here.

[0503] In step S3802, the model identifier B, the first data, and the eighth information are received.

[0504] The optional implementation of step S3802 can be found in the optional implementation of step S2406 in Figure 2D and other related parts in the embodiments involved in Figure 2D, which will not be repeated here.

[0505] In step S3803, based on model identifier B and model pairing relationship, a second model paired with the first model is determined.

[0506] The optional implementation of step S3803 can be found in the optional implementation of step S2407 in Figure 2D and other related parts in the embodiments involved in Figure 2D, which will not be repeated here.

[0507] In step S3804, the first data is processed using the second model.

[0508] The optional implementation of step S3804 can be found in the optional implementation of step S2408 in Figure 2D and other related parts in the embodiments involved in Figure 2D, which will not be repeated here.

[0509] Figure 4A is a flowchart illustrating the execution of a communication method on the first node side according to an embodiment of the present disclosure. As shown in Figure 4A, the embodiment of the present disclosure relates to a communication method executed by a first node. The communication method includes steps S4101 to S4102.

[0510] In step S4101, first information is obtained.

[0511] The optional implementations of step S4101 can be found in the optional implementations of step S2105 in Figure 2A, step S2204 in Figure 2B, step S2304 in Figure 2C, step S2403 in Figure 2D, other related parts in the embodiments involved in Figure 2A, other related parts in the embodiments involved in Figure 2B, other related parts in the embodiments involved in Figure 2C, and other related parts in the embodiments involved in Figure 2D. They will not be repeated here.

[0512] In step S4102, the model pairing relationship is determined based on the first information.

[0513] The optional implementations of step S4102 can be found in the optional implementations of step S2106 in Figure 2A, step S2205 in Figure 2B, step S2305 in Figure 2C, step S2404 in Figure 2D, other related parts in the embodiments involved in Figure 2A, Figure 2B, Figure 2C, and Figure 2D, and will not be repeated here.

[0514] Figure 4B is a flowchart illustrating the execution of a communication method on the second node side according to an embodiment of the present disclosure. As shown in Figure 4B, this embodiment of the disclosure relates to a communication method executed by a second node. The communication method includes step S4201.

[0515] In step S4201, first information is configured for the first model.

[0516] The optional implementations of step S4201 can be found in the optional implementations of step S2104 in Figure 2A, step S2203 in Figure 2B, step S2304 in Figure 2C, step S2403 in Figure 2D, other related parts in the embodiments involved in Figure 2A, other related parts in the embodiments involved in Figure 2B, other related parts in the embodiments involved in Figure 2C, and other related parts in the embodiments involved in Figure 2D. They will not be repeated here.

[0517] Figure 4C is a flowchart illustrating the execution of a communication method on the third node side according to an embodiment of the present disclosure. As shown in Figure 4C, this embodiment of the disclosure relates to a communication method executed by a third node. The communication method includes steps S4301 to S4303.

[0518] In step S4301, the first model is trained based on the third model and / or the second information associated with the third model.

[0519] The optional implementations of step S4301 can be found in the optional implementations of step S2102 in Figure 2A, step S2202 in Figure 2B, step S2302 in Figure 2C, step S2402 in Figure 2D, other related parts in the embodiments involved in Figure 2A, other related parts in the embodiments involved in Figure 2B, other related parts in the embodiments involved in Figure 2C, and other related parts in the embodiments involved in Figure 2D. They will not be repeated here.

[0520] In step S4302, first information for indicating the first model is obtained.

[0521] The optional implementations of step S4302 can be found in the optional implementations of step S2104 in Figure 2A, step S2203 in Figure 2B, step S2304 in Figure 2C, step S2403 in Figure 2D, other related parts in the embodiments involved in Figure 2A, other related parts in the embodiments involved in Figure 2B, and other related parts in the embodiments involved in Figure 2C, which will not be repeated here.

[0522] In step S4303, the first information is sent to the first node.

[0523] The optional implementations of step S4303 can be found in the optional implementations of step S2105 in Figure 2A, the optional implementations of step S2204 in Figure 2B, other related parts in the embodiments involved in Figure 2A, other related parts in the embodiments involved in Figure 2B, and other related parts in the embodiments involved in Figure 2C, which will not be repeated here.

[0524] Figure 4D is a flowchart illustrating the communication method executed by the fourth node according to an embodiment of the present disclosure. As shown in Figure 4D, the embodiment of the present disclosure relates to a communication method executed by the fourth node. The communication method includes step S4401.

[0525] In step S4401, the third model and / or the second information associated with the third model are sent.

[0526] The optional implementations of step S4401 can be found in the optional implementations of step S2101 in Figure 2A, step S2201 in Figure 2B, step S2301 in Figure 2C, step S2401 in Figure 2D, other related parts in the embodiments involved in Figure 2A, Figure 2B, Figure 2C, and Figure 2D, and will not be repeated here.

[0527] Figure 4E is a flowchart illustrating the communication method executed on the fifth node side according to an embodiment of the present disclosure. As shown in Figure 4E, this embodiment of the present disclosure relates to a communication method executed by a fifth node. The communication method includes steps S4501 to S4502.

[0528] In step S4501, first information and first data processed by the first model are received.

[0529] The optional implementations of step S4501 can be found in the optional implementations of step S2108 in Figure 2A, step S2207 in Figure 2B, step S2307 in Figure 2C, step S2406 in Figure 2D, other related parts in the embodiments involved in Figure 2A, Figure 2B, Figure 2C, and Figure 2D, and will not be repeated here.

[0530] In step S4502, the first data is processed using a second model paired with the first model.

[0531] The optional implementations of step S4501 can be found in the optional implementations of step S2110 in Figure 2A, step S2209 in Figure 2B, step S2308 in Figure 2C, step S2408 in Figure 2D, other related parts in the embodiments involved in Figure 2A, Figure 2B, Figure 2C, and Figure 2D, and will not be repeated here.

[0532] In some embodiments, the embodiments of the first node in FIG4A, the second node in FIG4B, the third node in FIG4C, and the fourth node in FIG4D can be used in combination or implemented separately.

[0533] In the following, the technical solutions of the embodiments of this disclosure will be described by way of specific implementation.

[0534] In some embodiments, for case 1, the network device transmits at least one of the model, training data, and model parameters to the terminal.

[0535] In some embodiments, the terminal side retrains an encoder model based on at least one of the model, training data, and model parameters.

[0536] In some embodiments, a retrained encoder is assigned at least one of encoder ID, paired ID, dataset ID, and training session ID, and model pairing is achieved based on the assigned encoder ID, paired ID, dataset ID, and training session ID.

[0537] In some embodiments, the ID assigned to the retrained encoder ID is the same as the encoder ID, paired ID, dataset ID, and training session ID determined before training.

[0538] In some embodiments, the network device transmits configuration information associated with the collection of training datasets or condition information from the network device to the terminal, and performs model pairing based on the associated configuration information or condition information.

[0539] In some embodiments, configuration information or condition information is distinguished by defining an associated ID.

[0540] In some embodiments, the configuration information described above may include the number of ports, the data type of the model input, the quantization method, etc.; the condition information on the network device side may include the channel scenario, etc.

[0541] In some embodiments, model pairing is achieved by using one or more of the following IDs: encoder ID, paired ID, dataset ID, and training session ID, as well as one or more of the following information: configuration information, network device-side condition information, and associated ID.

[0542] In some embodiments, the encoder ID, paired ID, dataset ID, training session ID, and associated ID are determined in at least one of the following ways:

[0543] Method 1: The network device assigns and instructs the terminal.

[0544] Method 1-1: One or more of the following IDs are transmitted to the terminal by the network device: encoder ID, paired ID, dataset ID, training session ID, associated ID, and at least one of the following: model, dataset, and model parameters.

[0545] Method 1-2: After the model is updated, the terminal indicates to the network device that the model retraining is complete, and then the network device assigns one or more of the following IDs to the terminal: encoder ID, paired ID, dataset ID, training session ID, and associated ID.

[0546] Method 2: The terminal determines and reports to the network device.

[0547] Method 2-1: After receiving one or more of the following from the network device: model, model parameters, dataset, configuration information, and condition information from the network device side, the terminal reports one or more of the following IDs to the network device: encoder ID, paired ID, dataset ID, training session ID, and associated ID.

[0548] Method 2-2: After the terminal completes model training, it reports one or more of the following IDs to the network device: encoder ID, paired ID, dataset ID, training session ID, and associated ID.

[0549] Method 3: Based on one or more of the predefined encoder ID, paired ID, dataset ID, training session ID, and associated ID, such as Encoder ID+X, where the value of X is a predefined related value, or is determined by network device indication or terminal reporting.

[0550] In some embodiments, for case 2, the terminal side transmits the model, dataset, or model parameters to the network device.

[0551] In some embodiments, a decoder model is retrained on the network device side.

[0552] In some embodiments, the model pairing method is similar to the model pairing method in Case 1 above.

[0553] In some embodiments, Case 2 differs from Case 1 in that a decoder ID is assigned to the retrained decoder.

[0554] In some embodiments, Case 2 differs from Case 1 in that the terminal sends terminal-side conditional information related to the training data to the network device. This terminal-side conditional information may include the maximum supported rank, terminal-side mobility state information, etc.

[0555] In some embodiments, the encoder ID, pair ID, dataset ID, and ID associated with configuration information or condition information are determined in a manner that includes at least one of the following:

[0556] Method 4: Network device allocation.

[0557] Method 4-1: After receiving one or more of the information such as model, dataset, and model parameters transmitted from the terminal side, the network device assigns one or more of the following IDs: decoder ID, paired ID, dataset ID, training session ID, and associated ID.

[0558] Method 4-2: After the network device completes the model update, it then assigns one or more of the following IDs: encoder ID, paired ID, dataset ID, training session ID, and associated ID.

[0559] Method 5: The terminal determines and sends the information to the network device.

[0560] Method 5-1: The terminal reports one or more of the following information to the network device: the model, model parameters, dataset, and terminal-side condition information, along with one or more of the following IDs: encoder ID, paired ID, dataset ID, training session ID, and associated ID.

[0561] Method 5-2: After the network device completes the Decoder model training, it notifies the terminal that model training is complete via an indication message. Upon receiving this indication message, the terminal reports one or more of the following IDs to the network device: encoder ID, paired ID, dataset ID, training session ID, and associated ID.

[0562] Method 6: Determined based on a predefined method. Same as Method 3 in Case 1.

[0563] In some embodiments, for case 1, the encoder and decoder models have been trained or deployed on the network device side. The encoder and decoder on the network device side can also be standardized models. The DNcoder and decoder trained or deployed on the network device side serve as reference models.

[0564] In some embodiments, the network device sends the trained reference model dncoder to the terminal via model passing, model parameter passing, or the encoder input and output datasets during encoder training.

[0565] In some embodiments, the terminal side may train a new encoder based on the received model (which may include an encoder and / or a decoder).

[0566] In some embodiments, if the terminal has already obtained the model structure information of the encoder and / or decoder, the terminal can train a new encoder based on the received model parameters (the model may include the model parameters of the encoder and / or decoder).

[0567] In some embodiments, the terminal side may also train a new encoder based on the input and output datasets of the received encoder.

[0568] In some embodiments, the terminal side may include multiple different encoders. During inference, in order to ensure that the encoder used on the terminal side and the decoder used on the network device side are paired and matched, the following provides relevant methods to ensure that the encoder and decoder are matched.

[0569] In some embodiments, a terminal is assigned at least one of the following IDs: encoder ID, paired ID, dataset ID, and training session ID. For example, after the terminal trains a new encoder, the network device assigns a new encoder ID to the terminal. The network device maintains a mapping relationship between the newly trained encoder and the network device's decoder. When the terminal notifies the network device of the encoder it is using, the network device selects the corresponding decoder for inference. When deciding which encoder to use for inference, the terminal sends the ID of the selected encoder to the network device to indicate the encoder being used.

[0570] In some embodiments, if the newly trained encoder of the terminal still uses the original encoder ID, the network device will not assign a new ID to the terminal. Even if the encoder ID applies to multiple encoders, the network device will still use the decoder corresponding to that encoder for inference.

[0571] In some embodiments, if the network device only transmits the model or model parameters, the terminal still needs a corresponding training dataset to train the encoder. The terminal can collect the training dataset based on the network device's configuration information, such as the number of transmit antenna ports and transmission bandwidth, or it can collect the training dataset based on the network device's condition information, such as UMA or UMi scene information, sent by the receiving network device.

[0572] In some embodiments, the corresponding configuration information or condition information may be associated with an ID, namely an associated ID. This associated ID may be indicated to the terminal by the network device. The terminal associates the associated ID with the trained Encoder. If the terminal notifies the network device of the associated ID, the network device can determine the model used by the terminal based on this associated ID.

[0573] In some embodiments, if the ID of the encoder retrained by the terminal is the same as the original encoder ID passed to the terminal by the network device, the associated ID mentioned above can be used to distinguish between the two encoders. That is, the associated IDs corresponding to the two encoders are different. The network device can determine the encoder used by the terminal based on the associated ID of the encoder, and thus select the corresponding decoder.

[0574] In some embodiments, the above is merely an illustrative indication that the encoder ID or associated ID is assigned and indicated to the terminal by the network device. The encoder ID or associated ID may also be determined based on the terminal side or based on a protocol predefined method.

[0575] In some embodiments, in addition to the encoder ID and associated ID, it may also be one or more of the assigned paired ID, dataset ID, and training session ID. These IDs can be sent to the terminal along with the model, model parameters, and dataset, or they can be indicated to the terminal after model training is completed. The above-mentioned method of indicating IDs can be implemented through one or more of the following signaling methods: RRC, MAC-CE, and DCI.

[0576] In some embodiments, the assigned ID can be a global ID or a local ID. It's also possible that a single global ID contains multiple local IDs. For example, the encoder transmitted by the network device is associated with a local ID 1 under a global ID. After the terminal retrains an encoder, the network device indicates a local ID 2 under the same global ID to the terminal. This avoids assigning multiple global IDs to a single terminal.

[0577] In some embodiments, for case 2, the terminal first trains both the encoder and decoder models. Similar to case 1, the terminal sends the corresponding model, model parameters, or dataset to the network device, which then retrains a decoder. The dataset sent by the terminal to the network device serves as the input and output datasets for training the decoder. The network device can then train a new decoder based on the received model, model parameters, or dataset. To ensure that the decoder used by the network device matches the encoder used by the terminal, and to avoid performance loss due to model mismatch, both the terminal and network device need to achieve bilateral model matching using one or more of the following IDs: encoder ID, paired ID, dataset ID, training session ID, or associated ID.

[0578] In some embodiments, the associated ID is associated with terminal-side condition information such as the supported maximum rank and terminal-side mobility status information. When multiple decoders are deployed on the network device side, the corresponding decoder can be selected through the associated ID indicated by the terminal.

[0579] In some embodiments, the encoder ID, paired ID, dataset ID, training session ID, or associated ID may be assigned to the terminal by the network device, or reported by the terminal to the network device, or predefined. The encoder ID, paired ID, dataset ID, training session ID, or associated ID may be transmitted along with model parameters, the model, or the dataset, or transmitted or indicated to the peer after the model is updated.

[0580] This disclosure also proposes an apparatus for implementing any of the above methods. For example, a terminal is proposed, which includes units or modules for implementing the steps performed by the terminal in any of the above methods. Furthermore, another network device is proposed, including units or modules for implementing the steps performed by the network device (e.g., access network device, core network functional node, core network device, etc.) in any of the above methods.

[0581] It should be understood that the division of units or modules in the above device is only a logical functional division. In actual implementation, they can be fully or partially integrated into a single physical entity, or they can be physically separated. Furthermore, the units or modules in the device can be implemented by a processor calling software: for example, the device includes a processor connected to a memory containing instructions. The processor calls the instructions stored in the memory to implement any of the above methods or to implement the functions of the units or modules in the above device. The processor can be, for example, a general-purpose processor, such as a Central Processing Unit (CPU) or a microprocessor, and the memory can be internal or external to the device. Alternatively, the units or modules in the device can be implemented in the form of hardware circuits. The functionality of some or all of the units or modules can be achieved through the design of these hardware circuits, which can be understood as one or more processors. For example, in one implementation, the hardware circuit is an application-specific integrated circuit (ASIC). The functionality of some or all of the units or modules is achieved through the design of the logical relationships between the components within the circuit. In another implementation, the hardware circuit can be implemented using a programmable logic device (PLD). Taking a field-programmable gate array (FPGA) as an example, it can include a large number of logic gates. The connection relationships between the logic gates are configured through configuration files, thereby achieving the functionality of some or all of the units or modules. All units or modules of the above device can be implemented entirely through processor-called software, entirely through hardware circuits, or partially through processor-called software with the remaining parts implemented through hardware circuits.

[0582] In this embodiment, the processor is a circuit with signal processing capabilities. In one implementation, the processor can be a circuit with instruction read and execute capabilities, such as a Central Processing Unit (CPU), a microprocessor, a graphics processing unit (GPU) (which can be understood as a microprocessor), or a digital signal processor (DSP). In another implementation, the processor can implement certain functions through the logical relationships of hardware circuits. The logical relationships of the aforementioned hardware circuits are fixed or reconfigurable. For example, the processor is a hardware circuit implemented using an application-specific integrated circuit (ASIC) or a programmable logic device (PLD), such as an FPGA. In a reconfigurable hardware circuit, the process of the processor loading a configuration document and configuring the hardware circuit 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 hardware circuits designed for artificial intelligence, which can be understood as ASICs, such as Neural Network Processing Units (NPUs), Tensor Processing Units (TPUs), and Deep Learning Processing Units (DPUs).

[0583] Figure 5A is a schematic diagram of the structure of the first node proposed in an embodiment of this disclosure. As shown in Figure 5A, the first node 5100 may include a first transceiver module 5101 and a first processing module 5102. In some embodiments, the first transceiver module 5101 is used to acquire first information, which is used to indicate a first model. The first processing module 5102 is used to determine a model pairing relationship based on the first information, which is used to indicate at least one model pair. Each model pair includes a first model and a second model. The first model is trained based on a third model and / or second information associated with the third model. In some embodiments, the first transceiver module 5101 is used to perform at least one of the communication steps (e.g., step S2105, but not limited thereto) performed by the first node in any of the above methods, which will not be described in detail here.

[0584] In some embodiments, the first node described above can be deployed on a terminal or on a network device.

[0585] Figure 5B is an exemplary structural diagram of a second node provided according to an embodiment of the present disclosure. As shown in Figure 5B, the second node 5200 may include a second processing module 5201. In some embodiments, the second processing module 5201 may be configured to configure first information for a first model, the first information being used to determine a model pairing relationship, the model pairing relationship being used to indicate at least one model pair, each model pair including a first model and a second model, the first model being trained based on a third model and / or second information associated with the third model. In some embodiments, the second node 5200 may also include a second transceiver model 5202. In some embodiments, the second transceiver model may be configured to send first information to a third node, the third node being used to train the first model based on the third model. In some embodiments, the second transceiver module 5202 may be configured to perform at least one of the communication steps (e.g., steps S2103, S2104, but not limited thereto) performed by the second node in any of the above methods, which will not be described in detail here.

[0586] In some embodiments, the second node described above can be deployed on a terminal or on a network device.

[0587] Figure 5C is a schematic diagram of the structure of the third node proposed in an embodiment of this disclosure. As shown in Figure 5C, the third node 5300 may include a third processing module 5301 and a third transceiver module 5302. In some embodiments, the third processing module 5301 is used to train a first model based on a third model and / or second information associated with the third model. The third transceiver module 5302 is used to send first information to the first node; the first information is used to trigger the first node to determine a model pairing relationship, the model pairing relationship is used to indicate at least one model pair, each model pair including a first model and a second model. In some embodiments, the above-described third transceiver module 5302 is used to perform at least one of the communication steps (e.g., S2103, but not limited thereto) performed by the third node in any of the above methods, which will not be described in detail here.

[0588] In some embodiments, the aforementioned third node can be deployed on a terminal or on a network device.

[0589] Figure 5D is an exemplary structural diagram of a fifth node provided according to an embodiment of the present disclosure. As shown in Figure 5D, the fifth node 5400 may include a fourth transceiver module 5401 and a fourth processing module 5402. In some embodiments, the fourth transceiver module 5401 may be configured to receive first information and first data processed by a first model. In some embodiments, the fourth processing module 5402 may be configured to process the first data using a second model paired with the first model; wherein the first information is used to indicate the first model, the second model is determined based on a model pairing relationship, the model pairing relationship is determined based on the first information, and the model pairing relationship is used to indicate at least one model pair, each model pair including a first model and a second model. In some embodiments, the fourth transceiver module 5401 may be configured to perform at least one of the communication steps such as sending and / or receiving performed by the fifth node in any of the above methods (e.g., step S2108, but not limited thereto), which will not be described in detail here.

[0590] In some embodiments, the aforementioned third node can be deployed on a terminal or on a network device.

[0591] In some embodiments, the transceiver module described above may include a transmitting module and / or a receiving module. The transmitting module and the receiving module may be separate or integrated together. Optionally, the transceiver module described above may be interchangeable with a transceiver.

[0592] Figure 6 is a schematic diagram of the structure of a communication device provided according to an embodiment of the present disclosure. The communication device 6100 can be any one of a first node, a second node, a third node, a fourth node, and a fifth node; it can also be a chip, chip system, or processor that supports the first node in implementing any of the above methods; it can also be a chip, chip system, or processor that supports the second node in implementing any of the above methods; it can also be a chip, chip system, or processor that supports the third node in implementing any of the above methods; it can also be a chip, chip system, or processor that supports the fourth node in implementing any of the above methods; and it can also be a chip, chip system, or processor that supports the fifth node in implementing any of the above methods. The communication device 6100 can be used to implement the methods described in the above method embodiments, and specific details can be found in the descriptions in the above method embodiments.

[0593] As shown in Figure 6, the communication device 6100 includes one or more processors 6101. The processor 6101 can be a general-purpose processor or a dedicated processor, such as a baseband processor or a central processing unit (CPU). The baseband processor can be used to process communication protocols and communication data, while the CPU can be used to control communication devices (e.g., base stations, baseband chips, terminal devices, terminal device chips, DUs or CUs, etc.), execute programs, and process program data. Optionally, the communication device 6100 can be used to execute any of the above methods. Optionally, one or more processors 6101 can be used to invoke instructions to cause the communication device 6100 to execute any of the above methods.

[0594] In some embodiments, the communication device 6100 further includes one or more transceivers 6102. When the communication device 6100 includes one or more transceivers 6102, the transceiver 6102 performs at least one of the communication steps such as sending and / or receiving in the above method (e.g., steps S3101, S3201, S3301, S3401, but not limited thereto), and the processor 6101 performs at least one of other steps (e.g., steps S3104, S3402, but not limited thereto). In optional embodiments, the transceiver 6102 may include a receiver and / or a transmitter, which may be separate or integrated together. Optionally, the terms transceiver, transceiver unit, transceiver, transceiver circuit, interface circuit, interface, etc., can be used interchangeably; the terms transmitter, transmitting unit, transmitter, transmitting circuit, etc., can be used interchangeably; and the terms receiver, receiving unit, receiver, receiving circuit, etc., can be used interchangeably.

[0595] In some embodiments, the communication device 6100 further includes one or more memories 6103 for storing data. Optionally, all or part of the memories 6103 may be located outside the communication device 6100. In optional embodiments, the communication device 6100 may include one or more interface circuits 6104. Optionally, the interface circuits 6104 are connected to the memories 6103 and can be used to receive data from the memories 6103 or other devices, and to send data to the memories 6103 or other devices. For example, the interface circuits 6104 can read data stored in the memories 6103 and send that data to the processor 6101.

[0596] The communication device 6100 described in the above embodiments may be a network device or a terminal, but the scope of the communication device 6100 described in this disclosure is not limited thereto, and the structure of the communication device 6100 may not be limited by FIG. 6. The communication device may be a standalone device or a part of a larger device. For example, the communication device may be: (1) a standalone integrated circuit IC, or chip, or chip system or subsystem; (2) a collection of one or more ICs, optionally, the IC collection may also include storage components 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, terminal device, smart terminal device, cellular phone, wireless device, handheld device, mobile unit, vehicle device, network device, cloud device, artificial intelligence device, etc.; (6) others, etc.

[0597] Figure 7 is a schematic diagram of the structure of a chip provided according to an embodiment of the present disclosure. For cases where the communication device 6100 can be a chip or a chip system, please refer to the schematic diagram of the chip 7100 shown in Figure 7, but it is not limited thereto.

[0598] Chip 7100 includes one or more processors 7101. Chip 7100 is used to perform any of the above methods.

[0599] In some embodiments, chip 7100 further includes one or more interface circuits 7102. Optionally, terms such as interface circuit, interface, and transceiver pin can be used interchangeably. In some embodiments, chip 7100 further includes one or more memories 7103 for storing data. Optionally, all or part of the memories 7103 may be located outside chip 7100. Optionally, interface circuit 7102 is connected to memory 7103, and interface circuit 7102 can be used to receive data from memory 7103 or other devices, and interface circuit 7102 can be used to send data to memory 7103 or other devices. For example, interface circuit 7102 can read data stored in memory 7103 and send the data to processor 7101.

[0600] In some embodiments, the interface circuit 7102 performs at least one of the communication steps such as sending and / or receiving in the above-described method (e.g., steps S3101, S3201, S3301, and S3401, but not limited thereto). The interface circuit 7102 performing the communication steps such as sending and / or receiving in the above-described method refers, for example, to the interface circuit 7102 performing data interaction between the processor 7101, the chip 7100, the memory 7103, or the transceiver device. In some embodiments, the processor 7101 performs at least one of other steps (e.g., steps S3104 and S3402, but not limited thereto).

[0601] The modules and / or devices described in the various embodiments, such as virtual devices, physical devices, and chips, can be combined or separated arbitrarily as needed. Optionally, some or all steps can also be performed collaboratively by multiple modules and / or devices, which is not limited here.

[0602] This disclosure also proposes a storage medium storing instructions that, when executed on a communication device 6100, cause the communication device 6100 to perform any of the methods described above. Optionally, the storage medium is an electronic storage medium. Optionally, the storage medium is a computer-readable storage medium, but not limited thereto; it may also be a storage medium readable by other devices. Optionally, the storage medium may be a non-transitory storage medium, but not limited thereto; it may also be a temporary storage medium.

[0603] This disclosure also provides a program product that, when executed by a communication device 6100, causes the communication device 6100 to perform any of the above methods. Optionally, the program product is a computer program product.

[0604] This disclosure also proposes a computer program that, when run on a computer, causes the computer to perform any of the above methods.

[0605] Other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This disclosure is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of the invention are indicated by the following claims.

[0606] It should be understood that the present invention is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims.

Claims

A communication method, executed by a first node, the method comprising: Obtain first information, which is used to instruct the first model; Based on the first information, a model pairing relationship is determined, which is used to indicate at least one model pair. Each model pair includes the first model and the second model. The first model is trained based on the third model and / or the second information associated with the third model. According to the method of claim 1, wherein, The first model is used to execute a first processing procedure, and the second model is used to execute a second processing procedure, wherein the first processing procedure and the second processing procedure are inverse processes of each other. The method according to claim 2, wherein, The first model and the third model are used to execute the first processing procedure. The third model is indicated by third information configured by the second node, and the first information is the same as the third information. The method according to claim 2, wherein, The second model and the third model are used to perform the second processing procedure, wherein the second model is the third model; or, the second model is trained based on the third model and / or the second information. The method according to any one of claims 1 to 4, wherein, The first information includes at least one of the following: first indication information for indicating the first model; second indication information for indicating the first model and the second model; and third indication information for indicating the training data of the first model. The fourth indication information is used to indicate the training session of the first model; The fifth instruction information is used to indicate the configuration of the type association of the training data of the first model; The sixth instruction information is used to indicate the conditions for collecting training data of the first model. The method according to any one of claims 1 to 5, wherein, The second information includes at least one of the following: third information for indicating the third model; fourth information for indicating the model structure of the third model; fifth information for indicating the model parameters of the third model; and sixth information for indicating the training data of the third model. The method according to any one of claims 1 to 6, wherein, The first information is determined based on the third information indicating the third model and the first numerical calculation. The method according to any one of claims 1 to 7, wherein, The first information is also used to trigger pairing of the first model. A communication method, executed by a second node, the method comprising: Configure first information for a first model, the first information being used to determine model pairing relationships, the model pairing relationships being used to indicate at least one model pair, each model pair including the first model and a second model, the first model being trained based on a third model and / or second information associated with the third model. The method according to claim 9, wherein, The first model is used to execute a first processing procedure, and the second model is used to execute a second processing procedure, wherein the first processing procedure and the second processing procedure are inverse processes of each other. The method according to claim 10, wherein, The first model and the third model are used to execute the first processing procedure. The third model is indicated by third information configured by the second node, and the first information is the same as the third information. The method according to claim 10, wherein, The second model and the third model are used to perform the second processing procedure, wherein the second model is the third model; or, the second model is trained based on the third model. The method according to any one of claims 9 to 12, wherein, The first information includes at least one of the following: first indication information for indicating the first model; second indication information for indicating the first model and the second model; and third indication information for indicating the training data of the first model. The fourth indication information is used to indicate the training session of the first model; The fifth instruction information is used to indicate the configuration of the type association of the training data of the first model; The sixth instruction information is used to indicate the conditions for collecting training data of the first model. The method according to any one of claims 9 to 13, wherein, The second information includes at least one of the following: third information for indicating the third model; fourth information for indicating the model structure of the third model; fifth information for indicating the model parameters of the third model; and sixth information for indicating the training data of the third model. The method according to any one of claims 9 to 14, wherein, The first information is determined based on the third information indicating the third model and the first numerical calculation. The method according to any one of claims 9 to 15, wherein, The method further includes: sending first information to a third node, the third node being used to train the first model based on the third model. The method according to claim 16, wherein, The method further includes: receiving a first message sent by a third node, wherein the first message is used to request the first information. A communication method, executed by a third node, the method comprising: The first model is trained based on the third model and / or the second information associated with the third model. Obtain first information used to indicate the first model; Send the first information to the first node; The first information is used to trigger the first node to determine the model pairing relationship, and the model pairing relationship is used to indicate at least one model pair, each model pair including the first model and the second model. The method according to claim 18, wherein, The first model is used to execute a first processing procedure, and the second model is used to execute a second processing procedure, wherein the first processing procedure and the second processing procedure are inverse processes of each other. The method according to claim 19, wherein, The second model and the third model are used to perform the second processing procedure, wherein the second model is the third model; or, the second model is trained based on the third model and / or the second information. The method according to claim 19, wherein, The first model and the third model are used to execute the first processing procedure. The third model is indicated by third information configured by the second node, and the first information is the same as the third information. The method according to any one of claims 18 to 20, wherein, The step of obtaining the first information associated with the first model includes: receiving the first information sent by the second node, wherein the second node is used to configure the first information for the first model. The method according to claim 22, wherein, The first information and the seventh information are received simultaneously. The second node and the fourth node are the same device. The fourth node is used to send the seventh information to the third node. The method according to claim 22, wherein, The method further includes: sending a first message to the second node, the first message being used to request the first information. The method according to any one of claims 18 to 24, wherein, The first information includes at least one of the following: first indication information for indicating the first model; second indication information for indicating the first model and the second model; and third indication information for indicating the training data of the first model. The fourth indication information is used to indicate the training session of the first model; The fifth instruction information is used to indicate the configuration of the type association of the training data of the first model; The sixth instruction information is used to indicate the conditions for collecting training data of the first model. The method according to any one of claims 18 to 25, wherein, The second information includes at least one of the following: third information for indicating the third model; fourth information for indicating the model structure of the third model; fifth information for indicating the model parameters of the third model; and sixth information for indicating the training data of the third model. The method according to any one of claims 18 to 26, wherein, The first information is determined based on the third information indicating the third model and the first numerical calculation. A communication method, executed by a fifth node, the method comprising: Receive the first information and the first data processed by the first model; The first data is processed using a second model paired with the first model; wherein the first information is used to indicate the first model, the second model is determined based on a model pairing relationship, the model pairing relationship is determined based on the first information, and the model pairing relationship is used to indicate at least one model pair, each model pair including the first model and the second model. The method according to claim 28, wherein, The first model is used to execute a first processing procedure, the second model is used to execute a second processing procedure, the first processing procedure and the second processing procedure are inverse processes of each other, and the first model is trained based on a third model and / or second information associated with the third model. The method according to claim 29, wherein, The first model and the third model are used to execute the first processing procedure. The third model is indicated by third information configured by the second node, and the first information is the same as the third information. The method according to claim 29, wherein, The second model and the third model are used to perform the second processing procedure, wherein the second model is the third model; or, the second model is trained based on the third model and / or the second information associated with the third model. The method according to any one of claims 28 to 31, wherein, The first information includes at least one of the following: first indication information for indicating the first model; second indication information for indicating the first model and the second model; and third indication information for indicating the training data of the first model. The fourth indication information is used to indicate the training session of the first model; The fifth instruction information is used to indicate the configuration of the type association of the training data of the first model; The sixth instruction information is used to indicate the conditions for collecting training data of the first model. The method according to any one of claims 29 to 32, wherein, The second information includes at least one of the following: third information for indicating the third model; fourth information for indicating the model structure of the third model; fifth information for indicating the model parameters of the third model; and sixth information for indicating the training data of the third model. The method according to any one of claims 28 to 33, wherein, The method further includes receiving seventh information, the seventh information being used to instruct the second model. The method according to any one of claims 28 to 34, wherein, The method further includes: receiving eighth information, the eighth information being used to indicate the model pairing relationship; and determining a second model paired with the first model based on the model pairing relationship. A first node includes: The first transceiver module is configured to acquire first information, which is used to indicate the first model. A first processing module is configured to determine a model pairing relationship based on the first information. The model pairing relationship is used to indicate at least one model pair, each model pair including the first model and the second model. The first model is trained based on the third model and / or the second information associated with the third model. A second node includes: The second processing module is configured to configure first information for the first model, the first information being used to determine model pairing relationships, the model pairing relationships being used to indicate at least one model pair, each model pair including the first model and the second model, the first model being trained based on the third model and / or the second information associated with the third model. A third node includes: The third processing module is configured to train the first model based on the third model and / or the second information associated with the third model. The third transceiver module is configured to send the first information to the first node; The first information is used to trigger the first node to determine the model pairing relationship, and the model pairing relationship is used to indicate at least one model pair, each model pair including the first model and the second model. A fifth node includes: The fourth transceiver module is configured to receive the first information and the first data processed by the first model; The fourth processing module is configured to process the first data using a second model paired with the first model; wherein the first information is used to indicate the first model, the second model is determined based on the model pairing relationship, the model pairing relationship is determined based on the first information, and the model pairing relationship is used to indicate at least one model pair, each model pair including the first model and the second model. A communication device, comprising: One or more processors; wherein the communication device is configured to perform the communication method according to any one of claims 1 to 35. A communication system includes a first node, a second node, a third node, a fourth node, and a fifth node; the first node is configured to implement the communication method as described in any one of claims 1 to 8; the second node is configured to implement the communication method as described in any one of claims 9 to 17; the third node is configured to implement the communication method as described in any one of claims 18 to 27; the fourth node is configured to send a third model and / or second information associated with the third model to the third node; and the fifth node is configured to implement the communication method as described in any one of claims 28 to 35. A storage medium storing instructions that, when executed on a communication device, cause the communication device to perform the communication method as described in any one of claims 1 to 35. A computer program product includes a computer program that, when executed by a processor, implements the communication method according to any one of claims 1 to 35.

Images