Communication method and related apparatus

Abstract

A communication method. In the method, first information acquired by a first communication apparatus is used for indicating inference state information of a first model, and the first communication apparatus can determine a second model on the basis of the inference state information of the first model. In this way, a first communication apparatus can perform model update on a model by means of inference state information of the model, so as to implement model management of the model. In addition, in the solution, the first communication apparatus can perform model update on a first model on the basis of inference state information of the first model, such that the inference performance of the updated model can be improved.

Description

A communication method and related apparatus

[0001] This application claims priority to Chinese Patent Application No. 202411458730.6, filed on October 17, 2024, entitled "A Communication Method and Related Device", the entire contents of which are incorporated herein by reference. Technical Field

[0002] This application relates to the field of communication technology, and in particular to a communication method and related apparatus. Background Technology

[0003] With the development of communication technology, communication equipment in communication systems can now perform not only traditional communication services but also other new types of services, such as artificial intelligence (AI) services. Generally, a communication system capable of handling AI services can also be called an AI system.

[0004] Currently, communication devices can serve as participating nodes in AI systems, applying their computing power to a specific stage of the AI ​​system. Generally, AI functions introduced into communication networks rely on models for implementation. However, in this process, there is currently no solution to address how to manage these models. Summary of the Invention

[0005] This application provides a communication method and related apparatus for implementing model management of model groups.

[0006] This application provides a communication method applied to a first communication device, for example, the method being executed by the first communication device. The first communication device may be a communication equipment (such as a terminal device or network device), or it may be a component of a communication equipment (such as a circuit or chip responsible for communication functions (such as a modem chip, also known as a baseband chip, or a system-on-chip (SoC) chip containing a modem core, or a system-in-package (SIP) chip), or it may be a logic module or software capable of implementing all or part of the functions of the communication equipment. In this method, the first communication device acquires first information, which is used to instruct updating (or processing, optimizing, iterating) a first model to a second model; wherein the first information is related to the inference state information of the first model (for example, the first information is determined based on the inference state information of the first model); the first communication device determines a second model based on the first information, the second model being different from the first model.

[0007] Based on the above scheme, the first information acquired by the first communication device is related to the inference state information of the first model, and the first communication device can determine the second model based on the update indication of the first information. In this way, the first communication device can update the model using the model's inference state information to achieve model management.

[0008] Furthermore, in the above scheme, the first communication device can update the first model based on the inference state information of the first model, which can improve the inference performance of the updated model.

[0009] In this application embodiment, the model may include an AI model, a mathematical model, a neural network model, an AI neural network model, a machine learning model, or an AI processing model, etc. Accordingly, the model involved in this application can be replaced with an AI model, a mathematical model, a neural network model, an AI neural network model, a machine learning model, or an AI processing model, etc.

[0010] In this embodiment of the application, a model is deployed on a communication device (e.g., a first model is deployed on a first communication device, a second model is deployed on a second communication device, etc.). This can be understood as follows: after the communication device obtains the model parameters of the model, it obtains, generates, or constructs the model based on the model parameters of the model. Subsequently, the communication device can adjust the model.

[0011] Optionally, model parameters may include one or more of the following: model hyperparameters, model dataset (including model input data and corresponding label data), model structure, model weights, and model parameters.

[0012] In the embodiments of this application, different communication devices (e.g., the first communication device and the second communication device) can transmit wireless communication signals (e.g., the transmission and reception of configuration information of communication resources, the transmission and reception of reference signals, etc.).

[0013] Optionally, the models involved in this application (e.g., the first model, the second model, and other models that may appear later) can be used to manage the wireless communication signal (including at least one of processing, configuration, updating, and optimization). For example, the model may include one or more of the following: a model for modulation and / or demodulation, a model for channel prediction, a model for beam management, a model for assisted positioning, a model for channel compression, a model for resource scheduling, and a model for replacing one or more modules in the transmitter and / or receiver. Alternatively, the models involved in this application may also be models for other tasks, such as models for image recognition, models for natural language processing, models for computer vision, etc.

[0014] Optionally, the inference state information of the first model mentioned above includes at least one of the following: the performance of the first model, the number of times the first model is used for model inference, or the duration of model inference using the first model, to improve the flexibility of the solution implementation. For example, the performance of the first model includes model performance (such as one or more of inference accuracy or inference latency) and / or system performance (such as one or more of throughput, packet loss rate, latency, or fairness).

[0015] Optionally, the inference state information can be replaced with other descriptions, such as inference performance information, inference information, model state information, or model performance information.

[0016] Optionally, reasoning can be replaced with other descriptions, such as prediction, deduction, identification, or decision.

[0017] In one possible implementation of the first aspect, the first communication device acquiring the first information includes: the first communication device receiving the first information.

[0018] Based on the above scheme, the first communication device can receive first information sent by other devices (such as the second communication device) to acquire the first information, so that the first communication device can update the first model based on the instructions of other devices.

[0019] For example, the first information mentioned above includes at least one of the following: the model identifier of the second model, or the model update method for updating the first model to the second model (e.g., model switching, model fine-tuning, or model training). In this way, the first communication device can update the first model based on the model identifier and / or model update method indicated by other devices, so that the first communication device can obtain the desired updated second model.

[0020] In one possible implementation of the first aspect, the method further includes: the first communication device sending second information indicating AI capability information of the first communication device; wherein the first information is determined based on the inference state information of the first model and the AI ​​capability information of the first communication device.

[0021] Based on the above scheme, the first communication device can send second information indicating the AI ​​capability information of the first communication device, so that the recipient of the second information can determine the first information based on the inference state information of the first model and the AI ​​capability information of the first communication device, and instruct the first communication device to update the first model through the first information.

[0022] Optionally, before sending the second information, the first communication device may receive an AI capability request message, which requests the AI ​​capability information of the first communication device. In this way, the first communication device can send its own AI capability information based on requests from other communication devices (such as the second communication device), so that the requesting party can obtain the AI ​​capability information of the first communication device and perform AI-related operations based on the AI ​​capability information of the first communication device (such as scheduling the first communication device to perform AI model inference and / or AI model updates, etc.).

[0023] Optionally, the above AI capability information indicates at least one of the following:

[0024] One or more supported models, wherein the one or more supported models include at least the first model;

[0025] Supported model update methods include at least one of model switching, model fine-tuning, or model training; or

[0026] Supported inference configuration methods, which indicate whether it is supported for inference configuration information to be specified by other devices, the inference configuration information including inference assistance information, or the inference configuration information including at least one of the following: model identifier.

[0027] Optionally, the inference configuration information includes one or more model identifiers and / or inference aid information. For example, the inference configuration information includes one or more model identifiers, which indicates that a model with the one or more model identifiers conforms to or matches the inference configuration. Alternatively, the inference configuration information includes inference aid information, which indicates that the model's update, optimization, or iteration process needs to be based on the inference aid information.

[0028] For example, inference aid information can indicate the configuration involved in the model inference process. For instance, the inference aid information may include one or more of the following: scene information (e.g., inference scene is indoor, outdoor, idle, busy, movement speed, etc.), model information (e.g., model performance, model complexity, etc.), environmental information (e.g., urban, suburban, etc.), system information (e.g., antenna configuration, system bandwidth, carrier frequency, subcarrier spacing, adoption frequency, throughput, packet loss rate, latency, fairness, etc.), or other information associated with the inference configuration.

[0029] In one possible implementation of the first aspect, the AI ​​capability information indicates a supported inference configuration method, the supported inference configuration method indicating support for inference configuration information specified by other devices; the method further includes: the first communication device receiving the inference configuration information, the inference configuration information being used to determine the second model.

[0030] Based on the above scheme, when the AI ​​capability information of the first communication device indicates that it supports the inference configuration information specified by other devices, the first communication device can also receive the inference configuration information, so that the first communication device can determine the second model based on the inference configuration information to obtain the second model that matches the inference configuration information, so that the subsequent inference process of the second model can obtain the expected inference result, thereby improving the inference performance.

[0031] In one possible implementation of the first aspect, the first communication device updates the first model to the second model based on the first information, including: the first communication device updates the first model to the second model based on the first information and inference configuration information.

[0032] Based on the above scheme, the first communication device can determine the second model based on the inference configuration information, thereby obtaining a second model that matches the inference configuration information. This enables the subsequent inference process of the second model to obtain the expected inference results, thus improving inference performance.

[0033] Optionally, the aforementioned inference configuration information may be provided by other communication devices, or the inference configuration information may be determined by the first communication device (e.g., pre-configured or pre-defined, or determined based on information it collects itself).

[0034] In one possible implementation of the first aspect, the first communication device acquiring first information includes: the first communication device acquiring inference state information of the first model; and the first communication device determining the first information based on the inference state information of the first model.

[0035] Based on the above scheme, the first communication device can determine the first information based on the inference state information of the first model after obtaining the inference state information of the first model, and update the first model based on the first information, so as to reduce overhead.

[0036] In one possible implementation of the first aspect, the method further includes: the first communication device sending third information, the third information indicating AI capability information of the first communication device; wherein the AI ​​capability information indicates a supported inference configuration method, the supported inference configuration method indicating support for inference configuration information specified by other devices; the method further includes: the first communication device receiving the inference configuration information, the inference configuration information being used to determine the second model.

[0037] Based on the above scheme, when the AI ​​capability information of the first communication device indicates that it supports the inference configuration information specified by other devices, the first communication device can also receive the inference configuration information, so that the first communication device can determine the second model based on the inference configuration information to obtain the second model that matches the inference configuration information, so that the subsequent inference process of the second model can obtain the expected inference result, thereby improving the inference performance.

[0038] Optionally, the AI ​​capability information of the first communication device may also refer to the implementation of the second information mentioned above. For example, the AI ​​capability information may also indicate one or more supported models and / or supported model update methods.

[0039] Optionally, before sending the third information, the first communication device may receive an AI capability request message, which requests the AI ​​capability information of the first communication device. In this way, the first communication device can send its own AI capability information based on requests from other communication devices (such as the second communication device), so that the requesting party can obtain the AI ​​capability information of the first communication device and perform AI-related operations based on the AI ​​capability information of the first communication device (such as scheduling the first communication device to perform AI model inference and / or AI model updates, etc.).

[0040] In one possible implementation of the first aspect, if the first condition is satisfied, the second model is one of one or more models supported by the first communication device.

[0041] Based on the above scheme, if the first condition is met, the second model is one of one or more models supported by the first communication device. That is, the first communication device can update the first model to the second model by switching models to achieve model update.

[0042] In this embodiment, the communication device can support one or more models, meaning it can deploy these models and support inference based on them. The model switching process described above can be understood as a communication device switching from one supported model to another.

[0043] As an example, the first communication device supports one or more models including model A and model B. Accordingly, in the above scheme, the first model can be model A and the second model can be model B; or, the first model can be model B and the second model can be model A.

[0044] As an example, the first condition is associated with the inference state information of the first model. Accordingly, the first condition is satisfied if at least one of the following is true: the performance of the first model is lower than or equal to a first threshold, the number of inferences of the first model is greater than or equal to a second threshold, or the inference duration of the first model is greater than or equal to a third threshold.

[0045] Optionally, the thresholds involved in this application (such as the first threshold, the second threshold, the third threshold, and other thresholds that may appear later) may be pre-configured or pre-defined, or may be configured by network devices or servers, and are not limited here.

[0046] In one possible implementation of the first aspect, if the first condition is satisfied and the fourth condition is satisfied, the second model is one of one or more models supported by the first communication device, the second model being adapted to the inference configuration information; wherein the fourth condition is associated with the inference configuration information.

[0047] Based on the above scheme, when both the first and fourth conditions are met, the second model is one of one or more models supported by the first communication device. That is, the first communication device can update the first model to the second model through model switching to achieve model updating. Furthermore, the updated second model is adapted to the inference configuration information, enabling the inference process of the second model to obtain the expected inference results, thereby improving inference performance.

[0048] For example, the fourth condition being met includes: at least one model among the models supported by the first communication device is compatible with the inference configuration information. Similarly, the fourth condition not being met includes: none of the models supported by the first communication device are compatible with the inference configuration information.

[0049] Taking the first communication device as an example, when the fourth condition is met, at least one model among the models supported by the first communication device is adapted to the inference configuration information; in other words, one or more models supported by the first communication device can be adapted to the inference configuration information, that is, the first model can be switched to any one of the one or more models, and the inference process of any one model can satisfy the inference configuration information.

[0050] Similarly, if the fourth condition is not met, none of the models supported by the first communication device are compatible with the inference configuration information; in other words, one or more models supported by the first communication device are not compatible with the inference configuration information, meaning that the first model cannot obtain a model that satisfies the inference configuration information through model switching.

[0051] In one possible implementation of the first aspect, the method further includes: the first communication device sending fourth information, the fourth information being used to instruct the first model to switch to the second model.

[0052] Based on the above scheme, the first communication device can also send a fourth message, enabling the first communication device to instruct or register the second model to the recipient of the fourth message, so that the recipient can subsequently schedule the second model obtained after model switching.

[0053] In one possible implementation of the first aspect, the second model is obtained by fine-tuning the first model when the second condition is met.

[0054] Based on the above scheme, when the second condition is met, the first communication device can update the first model to the second model by fine-tuning the model, thereby achieving model update.

[0055] In this embodiment of the application, the communication device can support model fine-tuning of one or more models. That is, after the communication device deploys one or more models, it can fine-tune the one or more models to obtain one or more fine-tuned models, and perform model inference based on the one or more fine-tuned models.

[0056] In the above process, the model fine-tuning process can be understood as: the first communication device fine-tunes a supported model to obtain the fine-tuned model.

[0057] For example, when a communication device fine-tunes a model, it can be understood that the device fine-tunes, adjusts, or updates some or all of the model's parameters. The number of parameters in this fine-tuning portion is relatively small (e.g., the number of parameters in this portion is below or equal to a threshold, or the ratio of the number of parameters in this portion to the total number of parameters in the model is below or equal to a threshold) and / or the number of adjustment rounds is relatively small (e.g., the number of adjustment rounds is less than or equal to a threshold). Optionally, model fine-tuning can be implemented online. For example, during model fine-tuning, the communication device can fine-tune, adjust, or update some parameters of the model online.

[0058] As an example, the second condition is associated with the inference state information of the first model. Accordingly, the second condition is satisfied if at least one of the following is true: the performance of the first model is lower than or equal to a fourth threshold, the number of inferences of the first model is greater than or equal to a fifth threshold, or the inference duration of the first model is greater than or equal to a sixth threshold.

[0059] Optionally, the fourth threshold may be less than or equal to the first threshold.

[0060] Optionally, the fifth threshold may be greater than or equal to the second threshold.

[0061] Optionally, the sixth threshold is greater than or equal to the third threshold.

[0062] In one possible implementation of the first aspect, if the second condition is satisfied and the fourth condition is not satisfied, the second model is obtained by fine-tuning the first model.

[0063] Based on the above scheme, if the second condition is met and the fourth condition is not met, the second model is obtained by fine-tuning the first model. That is, the first communication device can update the first model to the second model by fine-tuning the model to achieve the model update.

[0064] Optionally, the second model is adapted to the inference configuration information so that the inference process of the second model can obtain the desired inference results, thereby improving inference performance.

[0065] In one possible implementation of the first aspect, the method further includes: the first communication device sending fifth information, the fifth information being used to instruct the first model to obtain the second model through model fine-tuning.

[0066] Based on the above scheme, the first communication device can also send a fifth message, enabling the first communication device to instruct or register the second model to the recipient of the fifth message, so that the recipient can subsequently schedule the second model obtained after model fine-tuning.

[0067] In one possible implementation of the first aspect, the method further includes: the first communication device sending first request information, the first request information being used to request first data, the first data being used for model fine-tuning; and the first communication device receiving the first data.

[0068] Based on the above scheme, when the first communication device determines that the first model needs to be fine-tuned, the first communication device can send a first request message, so that the recipient of the first request message can send first data based on the first request message, so that the first communication device can achieve model fine-tuning of the first model based on the first data.

[0069] In one possible implementation of the first aspect, the second model is obtained by training the first model if the third condition is met.

[0070] Based on the above scheme, when the third condition is met, the first communication device can update the first model to the second model through model training to achieve model update.

[0071] In this embodiment of the application, the communication device can support model training of one or more models. That is, after the communication device deploys one or more models, it can train the one or more models to obtain one or more trained models, and perform model inference based on the one or more trained models.

[0072] In the above process, the model training process can be understood as: a communication device trains a model from a supported model to obtain the trained model.

[0073] For example, training a model using a communication device can be understood as the device training, adjusting, or updating some or all of the model's parameters. This involves a large number of parameters (e.g., the number of parameters is higher than or equal to a threshold, or the ratio of the number of parameters to the total number of parameters in the model is higher than or equal to a threshold) and / or a large number of adjustment rounds (e.g., the number of adjustment rounds is greater than or equal to a threshold). Optionally, model training can be implemented offline. For example, during model training, the communication device can train, adjust, or update some or all of the model's parameters offline.

[0074] As an example, the third condition is associated with the inference state information of the first model. Accordingly, the third condition is satisfied if at least one of the following is true: the performance of the first model is lower than or equal to the seventh threshold, the number of inferences of the first model is greater than or equal to the eighth threshold, or the inference duration of the first model is greater than or equal to the ninth threshold.

[0075] Optionally, the seventh threshold may be less than or equal to the fourth threshold.

[0076] Optionally, the eighth threshold is greater than or equal to the fifth threshold.

[0077] Optionally, the ninth threshold is greater than or equal to the sixth threshold.

[0078] In one possible implementation of the first aspect, if the third condition is satisfied and the fourth condition is not satisfied, the second model is obtained by training the model based on the first model.

[0079] Based on the above scheme, if the third condition is met and the fourth condition is not met, the second model is obtained by training the first model. That is, the first communication device can update the first model to the second model through model training to achieve model update.

[0080] Optionally, the second model is adapted to the inference configuration information so that the inference process of the second model can obtain the desired inference results, thereby improving inference performance.

[0081] In one possible implementation of the first aspect, the method further includes: the first communication device sending a sixth message, the sixth message being used to indicate that the first model has been trained to obtain the second model.

[0082] Based on the above scheme, the first communication device can also send a sixth message, enabling the first communication device to instruct or register the second model to the recipient of the sixth message, so that the recipient can subsequently schedule the second model obtained by the model training.

[0083] In one possible implementation of the first aspect, the method further includes: the first communication device sending a second request message for requesting second data, the first data being used for model training; and the first communication device receiving the second data.

[0084] Based on the above scheme, when the first communication device determines that it is training the first model, the first communication device can send a second request message, so that the recipient of the second request message can send second data based on the second request message, so that the first communication device can train the second model based on the second data.

[0085] The second aspect of this application provides a communication method applied to a second communication device, such as being executed by the second communication device, which may be a communication device (e.g., a terminal device or a network device), or the second communication device may be a component of the communication device (e.g., a circuit or chip responsible for communication functions (e.g., a modem chip, also known as a baseband chip, or a SoC chip or SIP chip containing a modem core, etc.), or the second communication device may also be a logic module or software capable of implementing all or part of the functions of the communication device.

[0086] In this method, the second communication device receives AI capability information from the first communication device (e.g., the AI ​​capability information of the first communication device is included in the second or third information mentioned above); wherein the AI ​​capability information indicates a supported inference configuration method, and the supported inference configuration method indicates whether it supports specifying inference configuration information by other devices; if the supported inference configuration method indicates that it supports specifying inference configuration information by other devices, the second communication device sends the inference configuration information, which is used to determine a second model; wherein the second model is obtained by updating the first model.

[0087] Based on the above scheme, the second communication device can receive AI capability information from the first communication device. When the AI ​​capability information indicates that the first communication device supports inference configuration information specified by other devices, the second communication device sends the inference configuration information to the first communication device. This allows the first communication device to update the first model based on the inference configuration information to obtain a second model. In this way, the first communication device can determine the second model based on the inference configuration information to obtain a second model that matches the inference configuration information. This ensures that the subsequent inference process of the second model yields the desired inference results, thereby improving inference performance.

[0088] In one possible implementation of the second aspect, the second model is obtained by updating the first model, including: the second model is obtained by updating the first model based on first information; wherein the first information is related to the inference state information of the first model (e.g., the first information is determined based on the inference state information of the first model), and the inference state information of the first model includes at least one of the following: the performance of the first model, the number of times model inference is performed using the first model, or the duration of model inference performed using the first model.

[0089] Based on the above scheme, the first communication device can determine the second model based on the inference state information of the first model. In this way, the first communication device can update the model using the model's inference state information, thereby achieving model management. Furthermore, in the above scheme, the first communication device can update the first model based on the inference state information of the first model, which can improve the inference performance of the updated model.

[0090] In one possible implementation of the second aspect, the method further includes: the second communication device sending the first information.

[0091] Based on the above scheme, the second communication device can also send first information to the first communication device, enabling the first communication device to update the first model based on the instructions of other devices.

[0092] Optionally, the above AI capability information indicates at least one of the following:

[0093] One or more supported models, wherein the supported models include at least the first model; or

[0094] Supported model update methods include at least one of model switching, model fine-tuning, or model training.

[0095] In one possible implementation of the second aspect, the second model satisfies any of the following:

[0096] If the first condition is met and the fourth condition is met, the second model is one of one or more models supported by the first communication device.

[0097] If the second condition is met and the fourth condition is not met, the second model is obtained by fine-tuning the first model; or

[0098] If the third condition is met and the fourth condition is not met, the second model is obtained by training the first model.

[0099] The fourth condition being met includes: one or more models supported by the first communication device contain a model that is compatible with the inference configuration information; the fourth condition not being met includes: one or more models supported by the first communication device do not contain a model that is compatible with the inference configuration information.

[0100] Based on the above scheme, when the above different conditions are met or not met, the first model can be updated by model switching, model fine-tuning, or model training, so that the updated second model can be adapted to the inference configuration information, thereby improving the inference performance of the second model in the future.

[0101] Optionally, at least one of the following must be satisfied:

[0102] The first condition includes at least one of the following: the performance of the first model is lower than or equal to a first threshold, the number of inferences of the first model is greater than or equal to a second threshold, or the inference time of the first model is greater than or equal to a third threshold;

[0103] The second condition includes at least one of the following: the performance of the first model is lower than or equal to the fourth threshold; the number of inferences of the first model is greater than or equal to the fifth threshold; or the inference duration of the first model is greater than or equal to the sixth threshold; or

[0104] The fourth condition includes at least one of the following: the performance of the first model is lower than or equal to the seventh threshold, the number of inferences of the first model is greater than or equal to the eighth threshold, or the inference duration of the first model is greater than or equal to the ninth threshold.

[0105] In the second aspect, the implementation behavior of the second communication device may also refer to the first aspect and related descriptions above, including but not limited to the process that the second communication device can receive at least one of the fourth information, the fifth information, and the sixth information, and achieve the corresponding technical effects.

[0106] A third aspect of this application provides a communication method applied to a second communication device, such as being executed by the second communication device, which may be a communication device (e.g., a terminal device or a network device), or the second communication device may be a component of the communication device (e.g., a circuit or chip responsible for communication functions (e.g., a modem chip, also known as a baseband chip, or a SoC chip or SIP chip containing a modem core, etc.), or the second communication device may also be a logic module or software capable of implementing all or part of the functions of the communication device.

[0107] In this method, a second communication device determines inference configuration information; the second communication device sends the inference configuration information; wherein the inference configuration information is used to determine the second model; wherein the second model is obtained by updating the first model.

[0108] Based on the above scheme, the second communication device can send inference configuration information to the first communication device, enabling the first communication device to update the first model to obtain the second model based on the inference configuration information. In this way, the first communication device can determine the second model based on the inference configuration information to obtain a second model that matches the inference configuration information, so that the subsequent inference process of the second model can obtain the expected inference results, thereby improving inference performance.

[0109] In the third aspect, the implementation behavior of the second communication device may also refer to the first or second aspect and related descriptions above, including but not limited to the process that the second communication device can receive at least one of the following: receiving second information, receiving third information, receiving fourth information, receiving fifth information, or receiving sixth information, and achieving the corresponding technical effect.

[0110] The fourth aspect of this application provides a communication method applied to a second communication device, such as being executed by the second communication device, which may be a communication device (e.g., a terminal device or a network device), or the second communication device may be a component of the communication device (e.g., a circuit or chip responsible for communication functions (e.g., a modem chip, also known as a baseband chip, or a SoC chip or SIP chip containing a modem core, etc.), or the second communication device may also be a logic module or software capable of implementing all or part of the functions of the communication device.

[0111] In this method, a second communication device determines first information, which is used to instruct the first model to be updated to a second model; wherein the first information is related to the inference state information of the first model, which includes at least one of the following: the performance of the first model, the number of times model inference is performed using the first model, or the duration of model inference performed using the first model; the second communication device sends the first information.

[0112] Based on the above scheme, the first information sent by the second communication device to the first communication device is related to the inference state information of the first model, and the first communication device can determine the second model based on the update instruction of the first information. In this way, the first communication device can update the model using the model's inference state information to achieve model management. Furthermore, in the above scheme, the first communication device can update the first model based on the first model's inference state information, which can improve the inference performance of the updated model.

[0113] In the fourth aspect, the implementation behavior of the second communication device may also refer to the first or second aspect and related descriptions above, including but not limited to the process that the second communication device can receive at least one of the following: receiving second information, receiving third information, receiving fourth information, receiving fifth information, or receiving sixth information, and achieving the corresponding technical effect.

[0114] A fifth aspect of this application provides a communication device including a processing unit; the processing unit is configured to acquire first information, the first information being configured to instruct a first model to be updated (or processed, optimized, iterated) to a second model; wherein the first information is related to the inference state information of the first model; the processing unit is further configured to determine a second model based on the inference state information of the first model, the second model being different from the first model.

[0115] In the fifth aspect of this application, the constituent modules of the communication device can also be used to perform the steps executed in various possible implementations of the first aspect and achieve the corresponding technical effects. For details, please refer to the first aspect, which will not be repeated here.

[0116] A sixth aspect of this application provides a communication device, which includes a transceiver unit and a processing unit. The transceiver unit is configured to receive AI capability information of a first communication device (e.g., the AI ​​capability information of the first communication device is included in the second or third information mentioned above). The AI ​​capability information indicates a supported inference configuration method, and the supported inference configuration method indicates whether it supports specifying inference configuration information by other devices. If the processing unit determines that the supported inference configuration method indicates support for specifying inference configuration information by other devices, the transceiver unit is further configured to send the inference configuration information, which is used to determine a second model. The second model is obtained by updating the first model.

[0117] In the sixth aspect of this application, the constituent modules of the communication device can also be used to perform the steps executed in various possible implementations of the second aspect and achieve the corresponding technical effects. For details, please refer to the second aspect, which will not be repeated here.

[0118] A seventh aspect of this application provides a communication device, which includes a processing unit and a transceiver unit; the processing unit is used to determine inference configuration information; the transceiver unit is used to transmit the inference configuration information; wherein the inference configuration information is used to determine a second model; wherein the second model is obtained by updating a first model.

[0119] In the seventh aspect of this application, the constituent modules of the communication device can also be used to perform the steps executed in various possible implementations of the third aspect and achieve the corresponding technical effects. For details, please refer to the third aspect, which will not be repeated here.

[0120] An eighth aspect of this application provides a communication device, the device including a processing unit and a transceiver unit; the processing unit is configured to determine first information, the first information being used to instruct updating a first model to a second model; wherein the first information is related to inference state information of the first model, the inference state information of the first model including at least one of the performance of the first model, the number of times model inference is performed using the first model, or the duration of model inference performed using the first model; the transceiver unit is configured to transmit the first information.

[0121] In the eighth aspect of this application, the constituent modules of the communication device can also be used to perform the steps executed in various possible implementations of the fourth aspect and achieve the corresponding technical effects. For details, please refer to the fourth aspect, which will not be repeated here.

[0122] The ninth aspect of this application provides a communication device including at least one processor for executing computer programs or instructions to enable the device to implement the method described in any one of the first to fourth aspects and any possible implementation thereof.

[0123] Optionally, the at least one memory is coupled to a memory used to store computer programs or instructions.

[0124] Optionally, the communication device includes the memory.

[0125] The tenth aspect of this application provides a communication device including at least one logic circuit and an input / output interface; the logic circuit is used to perform the method described in any of the possible implementations of the first to fourth aspects described above.

[0126] The eleventh aspect of this application provides a communication system, which includes the first communication device and the second communication device described above.

[0127] The twelfth aspect of this application provides a computer-readable storage medium for storing one or more computer-executable instructions, which, when executed by a processor, perform the method as described in any possible implementation of any of the first to fourth aspects described above.

[0128] The thirteenth aspect of this application provides a computer program product (or computer program) that, when executed by a processor, performs the method described in any possible implementation of any of the first to fourth aspects described above.

[0129] The fourteenth aspect of this application provides a chip or chip system including at least one processor for supporting a communication device in implementing the methods described in any possible implementation of any of the first to fourth aspects. For example, the chip may be a baseband chip, a modem chip, a system-on-a-chip (SoC) chip containing a modem core, a system-in-package (SIP) chip, or a communication module, etc.

[0130] In one possible design, the chip or chip system may further include a memory for storing program instructions and data necessary for the communication device. The chip system may be composed of chips or may include chips and other discrete devices. Optionally, the chip system may also include interface circuitry that provides program instructions and / or data to the at least one processor.

[0131] The technical effects of any of the design methods in aspects five through fourteen can be found in the technical effects of the different design methods in aspects one through four above, and will not be repeated here. Attached Figure Description

[0132] Figures 1a to 1c are schematic diagrams of the communication system provided in this application;

[0133] Figures 2a to 2g are schematic diagrams of the AI ​​processing involved in this application;

[0134] Figure 3 is an interactive schematic diagram of the communication method provided in this application;

[0135] Figures 4a to 4h are some schematic diagrams showing the application of the communication method provided in this application;

[0136] Figures 5 to 9 are schematic diagrams of the communication device provided in this application. Detailed Implementation

[0137] First, some terms used in the embodiments of this application will be explained to facilitate understanding by those skilled in the art.

[0138] (1) Terminal device: can be a wireless terminal device that can receive network device scheduling and instruction information. The wireless terminal device can be a device that provides voice and / or data connectivity to the user, or a handheld device with wireless connection function, or other processing device connected to a wireless modem.

[0139] Terminal devices can communicate with one or more core networks or the Internet via a radio access network (RAN). Terminal devices can be mobile terminal devices, such as mobile phones (or "cellular" phones), computers, and data cards. For example, they can be portable, pocket-sized, handheld, computer-embedded, or vehicle-mounted mobile devices that exchange voice and / or data with the RAN. Examples include personal communication service (PCS) phones, cordless phones, session initiation protocol (SIP) phones, wireless local loop (WLL) stations, personal digital assistants (PDAs), tablets, and computers with wireless transceiver capabilities. Wireless terminal equipment can also be referred to as a system, subscriber unit, subscriber station, mobile station (MS), remote station, access point (AP), remote terminal, access terminal, user terminal, user agent, subscriber station (SS), customer premises equipment (CPE), terminal, user equipment (UE), mobile terminal (MT), etc.

[0140] By way of example and not limitation, in this embodiment, the terminal device can also be a wearable device. Wearable devices, also known as wearable smart devices or smart wearable devices, are a general term for devices that utilize wearable technology to intelligently design and develop everyday wearables, such as glasses, gloves, watches, clothing, and shoes. Wearable devices are portable devices that are worn directly on the body or integrated into the user's clothing or accessories. Wearable devices are not merely hardware devices, but also achieve powerful functions through software support, data interaction, and cloud interaction. Broadly speaking, wearable smart devices include those that are feature-rich, large in size, and can achieve complete or partial functions without relying on a smartphone, such as smartwatches or smart glasses, as well as those that focus on a specific type of application function and require the use of other devices such as smartphones, such as various smart bracelets, smart helmets, and smart jewelry for vital sign monitoring.

[0141] Terminals can also be drones, robots, devices in device-to-device (D2D) communication, vehicles to everything (V2X) communication, virtual reality (VR) terminal devices, augmented reality (AR) terminal devices, wireless terminals in industrial control, wireless terminals in self-driving, wireless terminals in telemedicine or telehealth services, wireless terminals in smart grids, wireless terminals in transportation safety, wireless terminals in smart cities, wireless terminals in smart homes, etc.

[0142] Furthermore, the terminal device can also be a terminal device for a communication system evolved from the fifth generation (5G) communication system (such as 5G Advanced or future communication systems). For example, the form and function of the communication terminal can be further expanded, including but not limited to vehicles, cellular network terminals (integrating satellite terminal functions), drones, and Internet of Things (IoT) devices.

[0143] In this embodiment, the terminal device can also obtain artificial intelligence (AI) services provided by the network device. Optionally, the terminal device can also have AI processing capabilities.

[0144] (2) Network equipment: This can be equipment within a wireless network, such as RAN nodes (or devices) that connect terminal devices to the wireless network. Examples of RAN equipment currently include: base stations, evolved NodeBs (eNodeBs), gNBs (gNodeBs) in 5G communication systems, transmission reception points (TRPs), evolved Node Bs (eNBs), radio network controllers (RNCs), Node Bs (NBs), home base stations (e.g., home evolved Node Bs, or home Node Bs (HNBs), base band units (BBUs), or wireless fidelity (Wi-Fi) access points (APs). Additionally, in a network architecture, network equipment may include central unit (CU) nodes, distributed unit (DU) nodes, or RAN equipment comprising both CU and DU nodes.

[0145] Optionally, RAN nodes can also be macro base stations, micro base stations, indoor stations, relay nodes, donor nodes, or radio controllers in cloud radio access network (CRAN) scenarios. RAN nodes can also be servers, wearable devices, vehicles, or in-vehicle equipment. For example, the access network equipment in V2X technology can be a roadside unit (RSU).

[0146] In another possible scenario, multiple RAN nodes collaborate to assist the terminal in achieving wireless access, with different RAN nodes each implementing some of the base station's functions. For example, RAN nodes can be CUs, DUs, CUs (control plane, CP), CUs (user plane, UP), or radio units (RUs). CUs and DUs can be configured separately or included in the same network element, such as a baseband unit (BBU). RUs can be included in radio equipment or radio units, such as remote radio units (RRUs), active antenna units (AAUs), radio heads (RHs), or remote radio heads (RRHs).

[0147] In different systems, CU (or CU-CP and CU-UP), DU, or RU may have different names, but those skilled in the art will understand their meaning. For example, in an open access network (open RAN, O-RAN, or ORAN) system, CU can also be called O-CU (open CU), DU can also be called O-DU, CU-CP can also be called O-CU-CP, CU-UP can also be called O-CU-UP, and RU can also be called O-RU. For ease of description, this application uses CU, CU-CP, CU-UP, DU, and RU as examples. Any of the units among CU (or CU-CP, CU-UP), DU, and RU in this application can be implemented through software modules, hardware modules, or a combination of software modules and hardware modules.

[0148] Communication between access network devices and terminal devices follows a specific protocol layer structure. This protocol layer may include a control plane protocol layer and a user plane protocol layer. The control plane protocol layer may include at least one of the following: radio resource control (RRC) layer, packet data convergence protocol (PDCP) layer, radio link control (RLC) layer, media access control (MAC) layer, or physical (PHY) layer, etc. The user plane protocol layer may include at least one of the following: service data adaptation protocol (SDAP) layer, PDCP layer, RLC layer, MAC layer, or physical layer, etc.

[0149] The correspondence between network elements and their achievable protocol layer functions in the ORAN system can be found in Table 1 below.

[0150] Table 1

[0151] Network devices can be other devices that provide wireless communication functions for terminal devices. The embodiments of this application do not limit the specific technology or form of the network device. For ease of description, the embodiments of this application are not limited.

[0152] Network equipment may also include core network equipment, such as the Mobility Management Entity (MME), Home Subscriber Server (HSS), Serving Gateway (S-GW), Policy and Charging Rules Function (PCRF), and Public Data Network Gateway (PDN gateway or P-GW) in 4th generation (4G) networks; and access and mobility management function (AMF), user plane function (UPF), or session management function (SMF) in 5G networks. Furthermore, this core network equipment may also include other core network equipment in 5G networks and future networks.

[0153] In this embodiment of the application, the network device may also have network nodes with AI capabilities, which can provide AI services to terminals or other network devices. For example, it may be an AI node, computing node, RAN node with AI capabilities, or core network element with AI capabilities on the network side (access network or core network).

[0154] In this application embodiment, the device for implementing the function of the network device can be the network device itself, or it can be a device capable of supporting the network device in implementing the function, such as a chip system. This device can be disposed within the network device. In the technical solutions provided in this application embodiment, the example of a network device being used to implement the function of the network device is used to describe the technical solutions provided in this application embodiment.

[0155] (3) Configuration and Pre-configuration: In this application, both configuration and pre-configuration are used. Configuration refers to the network device / server sending configuration information or parameter values ​​to the terminal via messages or signaling, so that the terminal can determine communication parameters or resources for transmission based on these values ​​or information. Pre-configuration is similar to configuration; it can be parameter information or parameter values ​​pre-negotiated between the network device / server and the terminal device, parameter information or parameter values ​​specified by standard protocols for use by the base station / network device or terminal device, or parameter information or parameter values ​​pre-stored in the base station / server or terminal device. This application does not limit this.

[0156] Furthermore, these values ​​and parameters can be changed or updated.

[0157] (4) The terms "system" and "network" in the embodiments of this application can be used interchangeably. "Multiple" refers to two or more. "And / or" describes the relationship between related objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist simultaneously, or B exists alone, where A and B can be singular or plural. The character " / " generally indicates that the related objects before and after are in an "or" relationship. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, "at least one of A, B and C" includes A, B, C, AB, AC, BC or ABC. And, unless otherwise specified, the ordinal numbers such as "first" and "second" mentioned in the embodiments of this application are used to distinguish multiple objects and are not used to limit the order, sequence, priority or importance of multiple objects.

[0158] (5) In the embodiments of this application, "send" and "receive" indicate the direction of signal transmission. For example, "send information to XX" can be understood as the destination of the information being XX, which may include sending directly through the air interface or sending indirectly through the air interface by other units or modules. "Receive information from YY" can be understood as the source of the information being YY, which may include receiving directly from YY through the air interface or receiving indirectly from YY through the air interface by other units or modules. "Send" can also be understood as the "output" of the chip interface, and "receive" can also be understood as the "input" of the chip interface.

[0159] In the embodiments of this application, sending and receiving can be performed between devices, such as between a network device and a terminal device, or they can be performed within a device, such as between components, modules, chips, software modules, or hardware modules within the device via a bus, wiring, or interface.

[0160] In the embodiments of this application, information may undergo necessary processing, such as encoding and modulation, between the source and destination ends, but the destination end can understand the valid information from the source end. Similar expressions in this application can be understood in a similar way and will not be repeated here.

[0161] (6) In the embodiments of this application, "instruction" may include direct instruction and indirect instruction, as well as explicit instruction and implicit instruction. The information indicated by a certain piece of information (as described below, the instruction information) is called the information to be instructed. In the specific implementation process, there are many ways to indicate the information to be instructed, such as, but not limited to, directly indicating the information to be instructed, such as the information to be instructed itself or its index. It can also indirectly indicate the information to be instructed by indicating other information, where there is an association between the other information and the information to be instructed; or it can only indicate a part of the information to be instructed, while the other parts of the information to be instructed are known or pre-agreed upon. For example, the instruction of specific information can be achieved by using a pre-agreed (e.g., protocol predefined) arrangement order of various information, thereby reducing the instruction overhead to a certain extent. This application does not limit the specific method of instruction. In the embodiments of this application, for the sender of the instruction information, the instruction information can be used to indicate the information to be instructed; for the receiver of the instruction information, the instruction information can be used to determine the information to be instructed.

[0162] In this application, unless otherwise specified, the same or similar parts between the various embodiments can be referred to each other. In the various embodiments of this application, and the various methods / designs / implementations within each embodiment, unless otherwise specified or logically conflicting, the terminology and / or descriptions between different embodiments and between the various methods / designs / implementations within each embodiment are consistent and can be mutually referenced. The technical features in different embodiments and the various methods / designs / implementations within each embodiment can be combined to form new embodiments, methods, or implementations based on their inherent logical relationships. The following descriptions of the embodiments of this application do not constitute a limitation on the scope of protection of this application.

[0163] This application can be applied to long-term evolution (LTE) systems, new radio (NR) systems, or future communication systems beyond 5G. These communication systems include at least one network device and / or at least one terminal device.

[0164] Please refer to Figure 1a, which is a schematic diagram of the architecture of the communication system 1000 used in the embodiments of this application. As shown in Figure 1a, the communication system may include a Radio Access Network (RAN) 100. Optionally, the communication system 1000 may also include a core network 200 and an Internet 300. The RAN 100 includes at least one RAN node (110a and 110b in Figure 1a, collectively referred to as 110) and at least one terminal (120a-120j in Figure 1a, collectively referred to as 120). The RAN 100 may also include other RAN nodes, such as wireless relay devices and / or wireless backhaul devices (not shown in Figure 1a). The terminal 120 is wirelessly connected to the RAN node 110, and the RAN node 110 is wirelessly or wired connected to the core network 200. The core network equipment in the core network 200 and the RAN node 110 in the RAN 100 may be independent and different physical devices, or they may be the same physical device integrating the logical functions of the core network equipment and the logical functions of the RAN node. Terminals can be connected to each other, as can RAN nodes, via wired or wireless means.

[0165] Taking the communication system shown in Figure 1a as an example, in addition to performing communication-related services, different devices (including network devices and network devices, network devices and terminal devices, and / or terminal devices and terminal devices) may also perform AI-related services.

[0166] As shown in Figure 1b, taking a network device as a base station as an example, the base station can perform communication-related services and AI-related services with one or more terminal devices, and different terminal devices can also perform communication-related services and AI-related services.

[0167] As shown in Figure 1c, taking terminal devices including televisions and mobile phones as an example, communication-related services and AI-related services can also be performed between televisions and mobile phones.

[0168] The technical solutions provided in this application can be applied to wireless communication systems (such as the systems shown in Figures 1a, 1b, or 1c). For example, AI network elements can be introduced into the communication system provided in this application to realize some or all AI-related operations. AI network elements can also be called AI nodes, AI devices, AI entities, AI modules, AI models, or AI units, etc. The AI ​​network element can be built into a network element within the communication system. For example, the AI ​​network element can be an AI module built into: access network equipment, core network equipment, cloud server, or operation, administration, and maintenance (OAM) to realize AI-related functions. The OAM can act as the network management system for the core network equipment and / or the access network equipment. Alternatively, the AI ​​network element can also be an independently set network element in the communication system. Optionally, the terminal or its built-in chip can also include an AI entity to realize AI-related functions.

[0169] Optionally, in communication systems, AI application cases may include, but are not limited to: channel state information (CSI) prediction, beam management enhancement, positioning accuracy enhancement, network energy saving, load balancing, and mobility optimization. These will be explained below.

[0170] 1. CSI Prediction

[0171] For example, a terminal device reports channel quality information (CQI) to a network device. Based on the CQI, the network device selects an appropriate modulation and coding scheme (MCS) for the terminal device, thereby adapting to changing wireless channels. For instance, the terminal device performs channel estimation and prediction based on the received channel state information-reference signal (CSI-RS), and then feeds back the CQI to the network device. This fed-back CQI is used as input to the network device's model.

[0172] 2. Enhanced Beam Management

[0173] Enhanced beam management primarily aims to discover the strongest transmit / receive beam pairs. AI-based sparse beam prediction can improve accuracy. This can be achieved through both network-side and terminal-side AI sparse beam prediction, based on AI training and inference. Taking terminal-side AI sparse beam prediction as an example, the pre-trained AI model on the terminal device can be provided by the network or pre-stored on the terminal device. During training, the network device scans all possible beams and then provides the transmit beam pattern to the terminal device. Once training is complete, the network device only needs to scan a small subset of beams, and the terminal device then feeds back the inference results to the network device. AI-based beam management can achieve beam prediction in, for example, the temporal and / or spatial domains, reducing overhead and latency and improving beam selection accuracy.

[0174] Beam management enhancements may include at least one sub-function, such as beam scan matrix prediction and / or optimal beam prediction.

[0175] 3. Enhanced positioning accuracy

[0176] In line-of-sight (LOS) or non-line-of-sight (NLOS) scenarios, AI-based positioning can improve positioning accuracy with a smaller number of TRP antennas. Positioning enhancement can include at least one sub-function, such as: positioning enhancement based on access network devices, positioning enhancement based on positioning management function network elements, and positioning enhancement based on terminal devices.

[0177] 4. Network energy saving

[0178] Network energy conservation can be achieved through cell activation / deactivation, load reduction, coverage improvement, or other RAN setting adjustments. AI technology can be used to optimize energy-saving decisions by leveraging data collected in the RAN. AI algorithms can predict energy efficiency and load status for the next cycle, which can be used to assist in cell activation / deactivation decisions to save energy. Based on the predicted load, the system can dynamically configure energy-saving strategies to maintain a balance between system performance and energy efficiency, and reduce energy consumption.

[0179] 5. Load balancing

[0180] Load balancing can distribute the load evenly between cells and across different areas within a cell, or transfer some traffic from congested cells, or offload users across a single cell, carrier, or access standard, thereby improving network performance. Using AI models to enhance load balancing performance—such as inputting various measurements and feedback from terminal devices and network nodes, as well as historical data—can provide a higher quality user experience and increase system capacity.

[0181] 6. Mobility optimization or mobility management

[0182] Mobility management is a solution that ensures service continuity during terminal device mobility by minimizing dropped calls, radio link failures (RLFs), unnecessary handovers, and ping-pong effects. AI can enhance mobility management by, for example, reducing the probability of unexpected events, predicting at least one of the terminal device's location, mobility, or performance, and traffic redirection.

[0183] In the embodiments of this application, the definitions of the various technical terms mentioned above are merely illustrative. For example, as technology continues to develop, the scope of the above definitions may also change, and the embodiments of this application are not intended to limit the scope of these definitions.

[0184] For example, an AI function may include multiple AI sub-functions.

[0185] Optionally, AI application cases are also called AI application scenarios or AI functions.

[0186] As described above regarding AI application examples, AI can be widely used to improve network performance in areas such as CSI feedback enhancement, beam management, positioning accuracy enhancement, energy saving, mobility enhancement, and load balancing. AI models can typically be deployed on the network side and / or the terminal device side. The training of AI models relies on the collection of training data, which can come from measurements and feedback from the terminal devices.

[0187] The following is a brief introduction to the concepts that may be involved in this application.

[0188] AI can endow machines with some of the intelligence of humans; for example, it allows machines to use computer hardware and software to simulate certain intelligent human behaviors. To achieve artificial intelligence, machine learning methods can be employed. In machine learning, machines learn (or train) a model using training data. This model represents the mapping between input and output. The learned model can be used for reasoning (or prediction), that is, it can be used to predict the output corresponding to a given input. This output can also be called the reasoning result (or prediction result).

[0189] Machine learning can include supervised learning, unsupervised learning, and reinforcement learning. Unsupervised learning can also be called learning without supervision.

[0190] Supervised learning, based on collected sample values ​​and labels, uses machine learning algorithms to learn the mapping relationship between sample values ​​and labels, and then expresses this learned mapping relationship using an AI model. The process of training the machine learning model is the process of learning this mapping relationship. During training, sample values ​​are input into the model to obtain the model's predicted values, and the model parameters are optimized by calculating the error between the model's predicted values ​​and the sample labels (ideal values). After the mapping relationship is learned, it can be used to predict new sample labels. The mapping relationship learned in supervised learning can include linear or non-linear mappings. Based on the type of label, the learning task can be divided into classification tasks and regression tasks.

[0191] Unsupervised learning relies on collected sample values ​​to discover inherent patterns within the samples themselves. One type of unsupervised learning algorithm uses the samples themselves as supervisory signals, meaning the model learns the mapping relationship from sample to sample; this is called self-supervised learning. During training, model parameters are optimized by calculating the error between the model's predictions and the samples themselves. Self-supervised learning can be used for signal compression and decompression recovery applications; common algorithms include autoencoders and generative adversarial networks.

[0192] Reinforcement learning, unlike supervised learning, is a type of algorithm that learns problem-solving strategies through interaction with the environment. Unlike supervised and unsupervised learning, reinforcement learning problems do not have explicit "correct" action labels. The algorithm needs to interact with the environment to obtain reward signals from the environment, and then adjust its decision actions to obtain a larger reward signal value. For example, in downlink power control, the reinforcement learning model adjusts the downlink transmission power of each user based on the total system throughput feedback from the wireless network, aiming to achieve a higher system throughput. The goal of reinforcement learning is also to learn the mapping relationship between the environment state and a better (e.g., optimal) decision action. However, because the label of the "correct action" cannot be obtained in advance, the network cannot be optimized by calculating the error between the action and the "correct action." Reinforcement learning training is achieved through iterative interaction with the environment.

[0193] Neural networks (NNs) are a specific model in machine learning techniques. According to the general approximation theorem, neural networks can theoretically approximate any continuous function, thus enabling them to learn arbitrary mappings. Traditional communication systems rely on extensive expert knowledge to design communication modules, while deep learning communication systems based on neural networks can automatically discover hidden pattern structures from large datasets, establish mapping relationships between data, and achieve performance superior to traditional modeling methods.

[0194] The idea behind neural networks comes from the neuronal structure of the brain. For example, each neuron performs a weighted summation of its input values ​​and outputs the result through an activation function.

[0195] Figure 2a shows a schematic diagram of a neuron structure. Assume the neuron's input is x = [x0, x1, ..., x...]. n The weights corresponding to each input are w = [w0, w1, ..., w], respectively. n ], where n is a positive integer, w i and x i It can be any possible type, such as a decimal, an integer (e.g., 0, a positive integer, or a negative integer), or a complex number. i As x i The weights are used to assign weights to x. i Weighting is applied. The bias for the weighted sum of the input values ​​is, for example, b. Activation functions can take many forms. Suppose the activation function of a neuron is: y = f(z) = max(0, z), then the output of that neuron is: For example, if the activation function of a neuron is y = f(z) = z, then the output of that neuron is: Here, b can be any possible type, such as a decimal, an integer (e.g., 0, a positive integer, or a negative integer), or a complex number. The activation functions of different neurons in a neural network can be the same or different.

[0196] Furthermore, neural networks generally consist of multiple layers, each of which may include one or more neurons. Increasing the depth and / or width of a neural network can improve its expressive power, providing more powerful information extraction and abstract modeling capabilities for complex systems. The depth of a neural network can refer to the number of layers it includes, and the number of neurons in each layer can be called the width of that layer. In one implementation, a neural network includes an input layer and an output layer. The input layer processes the received input information through neurons and passes the processing result to the output layer, which then obtains the output of the neural network. In another implementation, a neural network includes an input layer, hidden layers, and an output layer. The input layer processes the received input information through neurons and passes the processing result to the hidden layer. The hidden layer calculates the received processing result and passes the calculation result to the output layer or the next adjacent hidden layer, ultimately obtaining the output of the neural network. A neural network may include one hidden layer or multiple sequentially connected hidden layers, without limitation.

[0197] Neural networks, for example, are deep neural networks (DNNs). Depending on how the network is constructed, DNNs can include feedforward neural networks (FNNs), convolutional neural networks (CNNs), and recurrent neural networks (RNNs).

[0198] Figure 2b is a schematic diagram of an FNN network. A characteristic of FNN networks is that neurons in adjacent layers are completely connected pairwise. This characteristic makes FNNs typically require a large amount of storage space, leading to high computational complexity.

[0199] CNNs are neural networks specifically designed to process data with a grid-like structure. For example, time-series data (e.g., discrete sampling along a time axis) and image data (e.g., two-dimensional discrete sampling) can both be considered grid-like data. CNNs do not use all the input information at once for computation; instead, they use a fixed-size window to extract a portion of the information for convolution operations, which significantly reduces the computational cost of model parameters. Furthermore, depending on the type of information extracted by the window (e.g., people and objects in an image represent different types of information), each window can use different convolution kernels, allowing CNNs to better extract features from the input data.

[0200] Recurrent Neural Networks (RNNs) are a type of neural network that utilizes feedback time-series information. The input to an RNN includes the current input value and its own output value from the previous time step. RNNs are suitable for acquiring temporally correlated sequence features, and are applicable to applications such as speech recognition and channel coding / decoding.

[0201] In the model training process described above, a loss function can be defined. The loss function describes the difference between the model's output value and the ideal target value. The loss function can be expressed in various forms, and there are no restrictions on its specific form. The model training process can be viewed as follows: by adjusting some or all of the model's parameters, the value of the loss function is made to be less than a threshold or to meet the target requirement.

[0202] A model can also be called an AI model, a rule, or other names. An AI model can be considered a specific method for implementing AI functions. An AI model represents the mapping relationship or function between the model's input and output. AI functions can include one or more of the following: data collection, model training (or model learning), model information dissemination, model inference (or model reasoning, inference, or prediction, etc.), model monitoring or model validation, or inference result publication, etc. AI functions can also be called AI (related) operations or AI-related functions.

[0203] The implementation process of the neural network will be described below with reference to the accompanying drawings.

[0204] Taking a fully connected neural network as an example, a fully connected neural network is also called a multilayer perceptron (MLP).

[0205] As shown in Figure 2c, an MLP consists of an input layer (left side), an output layer (right side), and multiple hidden layers (middle). Each layer of an MLP contains several nodes, called neurons. Neurons in adjacent layers are connected pairwise.

[0206] Optionally, considering neurons in two adjacent layers, the output h of the next layer's neurons is the weighted sum of all neurons x in the previous layer connected to it, processed by an activation function, and can be expressed as: h = f(wx + b).

[0207] Where w is the weight matrix, b is the bias vector, and f is the activation function.

[0208] Alternatively, the output of the neural network can be recursively expressed as: y = f z (w z f z-1 (…)+b z ).

[0209] Where z is the index of the neural network layer, z is greater than or equal to 1 and z is less than or equal to Z, where Z is the total number of layers in the neural network.

[0210] In this embodiment, a neural network can be understood as a mapping relationship from an input data set to an output data set. Typically, neural networks are randomly initialized, and the process of obtaining this mapping relationship from random values ​​w and b using existing data is called neural network training.

[0211] Optionally, the training method involves using a loss function to evaluate the output of the neural network.

[0212] As shown in Figure 2d, the error can be backpropagated, and the neural network parameters (including w and b) can be iteratively optimized using gradient descent until the output of the loss function reaches its minimum value, which is the "better point (e.g., the optimal point)" in Figure 2d. In this embodiment, the neural network parameters corresponding to the "better point (e.g., the optimal point)" in Figure 2d can be used as the neural network parameters in the trained AI model information.

[0213] Alternatively, the gradient descent process can be represented as:

[0214] Where θ represents the parameters to be optimized (including w and b), L is the loss function, and η is the learning rate, controlling the step size of gradient descent. This represents the differentiation operation. This indicates taking the derivative of θ with respect to L.

[0215] Alternatively, the backpropagation process can utilize the chain rule for partial derivatives.

[0216] As shown in Figure 2e, the gradient of the parameters in the previous layer can be recursively calculated from the gradient of the parameters in the next layer, and can be expressed as:

[0217] Among them, w ij Let s be the weight of the connection between node j and node i. i The weighted sum of the inputs at node i.

[0218] The technical solutions provided in this application can be applied to wireless communication systems (such as the systems shown in Figure 1a, 1b, or 1c). In wireless communication systems, communication nodes generally possess signal transmission and reception capabilities as well as computing capabilities. To address the vision of future intelligent and inclusive accessibility, intelligence will further evolve at the wireless network architecture level, taking AI and / or machine learning (ML) as examples. In future networks, AI and / or ML will be further deeply integrated with wireless networks to achieve network-inherent intelligence, which also includes the intelligence of terminals.

[0219] For example, neural network-based AI and / or ML can be applied to wireless communication systems. Through data-driven training, accurate modeling and prediction of wireless data can be achieved. For instance, AI and / or ML can be used for various wireless tasks, such as time-frequency domain channel estimation, prediction, and beam prediction related to wireless channels. AI and / or ML models for specific tasks require data collection, model training, selection, switching, and retraining based on applicable conditions.

[0220] As an example, Figure 2f illustrates how AI models can be deployed on one side of a wireless communication system, such as the UE side or the network side. For instance, an AI model for demodulation can be deployed on the receiver side; that is, during downlink transmission, the demodulation AI model is deployed on the UE side, and during uplink transmission, it is deployed on the network side. Similarly, an AI model for precoding can be deployed on the transmitter side. And again, an AI model for channel estimation can be deployed on the receiver side.

[0221] As an example, as shown in Figure 2g, the current 3GPP discussion provides a functional framework for using AI and / or ML to improve air interface performance, including data collection, model training, management, inference, and model storage. Optionally, these functions may interact with each other. For example, the data collection module can provide data to the model training module, management module, and inference module; the model training and management modules can provide models to the model storage module; and the model storage module can provide models for inference.

[0222] However, the above process only provides a framework of functional description. For communication devices, there is still no specific solution on how to manage models (especially models deployed on one side) (including model usage and / or updates).

[0223] To address the aforementioned problems, this application provides a communication method and related apparatus, which will be described in detail below with reference to the accompanying drawings.

[0224] Figure 3 is a schematic diagram of an implementation of the communication method provided in this application. Figure 3 illustrates the method using a first communication device and other communication devices (such as a second communication device) as examples of the execution subjects of this interactive illustration, but this application does not limit the execution subjects of this interactive illustration. For example, the communication device can be a communication equipment (such as a terminal device or network device), or a chip, baseband chip, modem chip, SoC chip (such as an SoC chip containing a modem core), SIP chip, communication module, chip system, processor, logic module, or software, etc., within the communication equipment.

[0225] As an example, the first communication device can be a terminal device and the second communication device can be a network device.

[0226] As another example, the first communication device can be a network device, and the second communication device can be a terminal device.

[0227] As another example, both the first and second communication devices are network devices.

[0228] Optionally, the aforementioned network equipment may be access network equipment or ORAN equipment (including at least one of O-CU, O-DU, and O-RU).

[0229] As another example, both the first and second communication devices are terminal devices, meaning that the scheme shown in Figure 3 can be applied to sidelink communication scenarios.

[0230] S301. The first communication device acquires first information, which is used to instruct the acquisition of a second model based on the first model, such as updating (or processing, optimizing, iterating) the first model to the second model; wherein, the first information is related to the inference state information of the first model.

[0231] S302. The first communication device determines a second model based on the first information, the second model being different from the first model.

[0232] In this application, the model may include an AI model, a mathematical model, a neural network model, an AI neural network model, a machine learning model, or an AI processing model, etc. Accordingly, the model involved in this application can be replaced with an AI model, a mathematical model, a neural network model, an AI neural network model, a machine learning model, or an AI processing model, etc.

[0233] In this application, a model is deployed on a communication device (e.g., a first model is deployed on a first communication device, a second model is deployed on a second communication device, etc.). For example, if the first model is deployed on the first communication device, the first communication device obtains the model parameters of the first model, and obtains, generates, or constructs the first model based on the model parameters. Subsequently, the first communication device can adjust the model of the first model.

[0234] Optionally, model parameters may include one or more of the following: model hyperparameters, model dataset (including model input data and corresponding label data), model structure, model weights, and model parameters.

[0235] In the embodiments of this application, different communication devices (e.g., the first communication device and the second communication device) can transmit wireless communication signals (e.g., the transmission and reception of configuration information of communication resources, the transmission and reception of reference signals, etc.).

[0236] Optionally, the models involved in this application (e.g., the first model, the second model, and other models that may appear later) can be used to manage the wireless communication signal (including at least one of processing, configuration, updating, and optimization). For example, the model may include one or more of the following: a model for modulation and / or demodulation, a model for channel prediction, a model for beam management, a model for assisted positioning, a model for channel compression, a model for resource scheduling, and a model for replacing one or more modules in the transmitter and / or receiver. Alternatively, the models involved in this application may also be models for other tasks, such as models for image recognition, models for natural language processing, models for computer vision, etc.

[0237] Optionally, the inference state information of the first model mentioned above includes at least one of the following: the performance of the first model, the number of times the first model is used for model inference, or the duration of model inference using the first model, to improve the flexibility of the solution implementation. For example, the performance of the first model includes model performance (such as one or more of inference accuracy or inference latency) and / or system performance (such as one or more of throughput, packet loss rate, latency, or fairness).

[0238] Optionally, the inference state information can be replaced with other descriptions, such as inference performance information, inference information, model state information, or model performance information.

[0239] Optionally, reasoning can be replaced with other descriptions, such as prediction, deduction, identification, or decision.

[0240] Based on the scheme shown in Figure 3, the first information obtained by the first communication device in step S301 is related to the inference state information of the first model. The first communication device determines the second model based on the indication of the first information, for example, by updating the first model to obtain the second model. In this way, the first communication device can update the model using the model's inference state information to achieve model management.

[0241] Furthermore, in the above scheme, the first communication device can update the first model based on the inference state information of the first model, which can improve the inference performance of the updated model.

[0242] In one possible implementation of the method shown in Figure 3, the first communication device determines the second model based on the first information, including updating the first model to the second model based on the first information and inference configuration information. Thus, the first communication device can determine the second model based on the inference configuration information to obtain a second model that matches the inference configuration information, enabling subsequent inference processes of the second model to obtain the desired inference results, thereby improving inference performance.

[0243] Optionally, the aforementioned inference configuration information may be provided by other communication devices (e.g., the implementation process of step B below), or the inference configuration information may be determined by the first communication device (e.g., pre-configured or pre-defined, or determined based on information it collects).

[0244] Optionally, the first communication device may acquire the first information in a variety of ways in step S301. Some possible implementations will be described below in conjunction with examples.

[0245] In Method I, the process of the first communication device acquiring the first information in step S301 includes: the second communication device sending the first information, and correspondingly, the first communication device receiving the first information.

[0246] Thus, in mode I, the first communication device can receive first information sent by other devices (e.g., the second communication device) to acquire the first information, enabling the first communication device to update the first model based on the instructions of other devices.

[0247] For example, the first information mentioned above includes at least one of the following: the model identifier of the second model, or the model update method for updating the first model to the second model (e.g., model switching, model fine-tuning, or model training). In this way, the first communication device can update the first model based on the model identifier and / or model update method indicated by other devices, so that the first communication device can obtain the desired updated second model.

[0248] In one possible implementation of method I, the method further includes: the first communication device sending second information, and correspondingly, the second communication device receiving the second information. The second information indicates the AI ​​capability information of the first communication device in step A of Figure 3, and the first information is determined based on the inference state information of the first model and the AI ​​capability information of the first communication device. Thus, the first communication device can send the second information indicating its AI capability information, enabling the recipient of the second information to determine the first information based on the inference state information of the first model and the AI ​​capability information of the first communication device, and to instruct the first communication device to update the first model through the first information.

[0249] Optionally, before sending the second information, the first communication device may receive an AI capability request message, which requests the AI ​​capability information of the first communication device. In this way, the first communication device can send its own AI capability information based on requests from other communication devices (such as the second communication device), so that the requesting party can obtain the AI ​​capability information of the first communication device and perform AI-related operations based on the AI ​​capability information of the first communication device (such as scheduling the first communication device to perform AI model inference and / or AI model updates, etc.).

[0250] Optionally, the above AI capability information indicates at least one of the following:

[0251] One or more supported models, wherein the one or more supported models include at least the first model;

[0252] Supported model update methods include at least one of model switching, model fine-tuning, or model training; or

[0253] Supported inference configuration methods, which indicate whether it is supported for inference configuration information to be specified by other devices, the inference configuration information including inference assistance information, or the inference configuration information including at least one of the following: model identifier.

[0254] Optionally, the inference configuration information includes one or more model identifiers and / or inference aid information. For example, the inference configuration information includes one or more model identifiers, which indicates that a model with the one or more model identifiers conforms to or matches the inference configuration. Alternatively, the inference configuration information includes inference aid information, which indicates that the model's update, optimization, or iteration process needs to be based on the inference aid information.

[0255] For example, inference aid information can indicate the configuration involved in the model inference process. For instance, the inference aid information may include one or more of the following: scene information (e.g., inference scene is indoor, outdoor, idle, busy, movement speed, etc.), model information (e.g., model performance, model complexity, etc.), environmental information (e.g., urban, suburban, etc.), system information (e.g., antenna configuration, system bandwidth, carrier frequency, subcarrier spacing, adoption frequency, throughput, packet loss rate, latency, fairness, etc.), or other information associated with the inference configuration.

[0256] In one possible implementation of Method I, the AI ​​capability information in step A indicates a supported inference configuration method, which indicates support for specifying inference configuration information by other devices. The method further includes: a second communication device sending inference configuration information in step B, and correspondingly, the first communication device receiving the inference configuration information, which is used to determine the second model. Thus, when the AI ​​capability information of the first communication device indicates support for specifying inference configuration information by other devices, the first communication device can also receive the inference configuration information, enabling it to determine the second model based on the inference configuration information to obtain a second model matching the inference configuration information. This allows subsequent inference processes of the second model to obtain the desired inference results, thereby improving inference performance.

[0257] In Method II, the process of the first communication device acquiring the first information in step S301 includes: the first communication device acquiring the inference state information of the first model; and the first communication device determining the first information based on the inference state information of the first model.

[0258] Therefore, in Method II, after acquiring the inference state information of the first model, the first communication device can determine the first information based on the inference state information of the first model, and update the first model based on the first information, thereby reducing overhead.

[0259] In one possible implementation of Method II, the method further includes: the first communication device sending third information, the third information indicating AI capability information of the first communication device in step A; wherein the AI ​​capability information indicates a supported inference configuration method, the supported inference configuration method indicating support for inference configuration information specified by other devices; the method further includes: the second communication device sending inference configuration information in step B, and correspondingly, the first communication device receiving the inference configuration information, the inference configuration information being used to determine the second model.

[0260] Therefore, when the AI ​​capability information of the first communication device indicates that it supports the inference configuration information specified by other devices, the first communication device can also receive the inference configuration information, so that the first communication device can determine the second model based on the inference configuration information to obtain the second model that matches the inference configuration information, so that the subsequent inference process of the second model can obtain the desired inference result, thereby improving the inference performance.

[0261] Optionally, the AI ​​capability information of the first communication device may also refer to the implementation of the second information mentioned above. For example, the AI ​​capability information may also indicate one or more supported models and / or supported model update methods.

[0262] Optionally, before sending the third information, the first communication device may receive an AI capability request message, which requests the AI ​​capability information of the first communication device. In this way, the first communication device can send its own AI capability information based on requests from other communication devices (such as the second communication device), so that the requesting party can obtain the AI ​​capability information of the first communication device and perform AI-related operations based on the AI ​​capability information of the first communication device (such as scheduling the first communication device to perform AI model inference and / or AI model updates, etc.).

[0263] Optionally, the second model can be determined based on the first model. The second model can be determined in various ways, which will be described below with some examples.

[0264] In Example 1, the second model is determined through model switching.

[0265] In one possible implementation of Example 1, if the first condition is met, the second model is one of one or more models supported by the first communication device. That is, the first communication device can update the first model to the second model by switching models to achieve model update.

[0266] In this embodiment, the communication device can support one or more models, meaning it can deploy these models and support inference based on them. The model switching process described above can be understood as a communication device switching from one supported model to another.

[0267] As an example, the first communication device supports one or more models including model A and model B. Accordingly, in the above scheme, the first model can be model A and the second model can be model B; or, the first model can be model B and the second model can be model A.

[0268] As an example, the first condition is associated with the inference state information of the first model. Accordingly, the first condition is satisfied if at least one of the following is true: the performance of the first model is lower than or equal to a first threshold, the number of inferences of the first model is greater than or equal to a second threshold, or the inference duration of the first model is greater than or equal to a third threshold.

[0269] Optionally, the thresholds involved in the embodiments of this application (such as the first threshold, the second threshold, the third threshold, and other thresholds that may appear later) may be pre-configured or pre-defined, or may be configured by network devices or servers, and are not limited here.

[0270] In another possible implementation of Example 1, when both the first and fourth conditions are met, the second model is one of one or more models supported by the first communication device, and this second model is adapted to the inference configuration information, wherein the inference configuration information is associated with the fourth condition. Thus, the updated second model is adapted to the inference configuration information, enabling the inference process of the second model to obtain inference results that conform to the expectations of the inference configuration information, thereby improving inference performance.

[0271] For example, the fourth condition being met includes: at least one model among the models supported by the first communication device is compatible with the inference configuration information. Similarly, the fourth condition not being met includes: none of the models supported by the first communication device are compatible with the inference configuration information.

[0272] Taking the first communication device as an example, when the fourth condition is met, at least one model among the models supported by the first communication device is adapted to the inference configuration information; in other words, one or more other models supported by the first communication device besides the first model can be adapted to the inference configuration information, that is, the first model can be switched to any of the other one or more models, and the inference process of any model can satisfy the inference configuration information.

[0273] Similarly, if the fourth condition is not met, none of the models supported by the first communication device are compatible with the inference configuration information; in other words, one or more models supported by the first communication device are not compatible with the inference configuration information, meaning that the first model cannot obtain a model that satisfies the inference configuration information through model switching.

[0274] Optionally, the inference configuration information in this application may indicate the configuration involved in the model inference process. For example, the inference configuration information may include one or more of the following: scene information (e.g., inference scene is indoor, outdoor, idle, busy, etc.), model information (e.g., model performance, model complexity, etc.), environmental information (e.g., urban area, suburbs, etc.), system information (e.g., antenna configuration, throughput, packet loss rate, latency, fairness, etc.), or other information related to the inference configuration.

[0275] Optionally, in implementing Example 1, the method shown in Figure 3 also includes:

[0276] S303. The first communication device sends a fourth message, and correspondingly, the second communication device receives the fourth message. The fourth message is used to indicate that the first model has been switched to the second model.

[0277] Therefore, the first communication device can also send a fourth message, enabling the first communication device to instruct or register the second model to the recipient of the fourth message, so that the recipient can subsequently schedule the second model obtained after model switching.

[0278] Example 2: The second model is determined through model fine-tuning.

[0279] In one possible implementation of Example 2, when the second condition is met, the second model is obtained by fine-tuning the first model. Thus, the first communication device can update the first model to the second model through model fine-tuning, thereby achieving model updating.

[0280] In this embodiment of the application, the communication device can support model fine-tuning of one or more models. That is, after the communication device deploys one or more models, it can fine-tune the one or more models to obtain one or more fine-tuned models, and perform model inference based on the one or more fine-tuned models.

[0281] In the above process, the model fine-tuning process can be understood as: the first communication device fine-tunes a supported model to obtain the fine-tuned model.

[0282] For example, when a communication device fine-tunes a model, it can be understood that the device fine-tunes, adjusts, or updates some or all of the model's parameters. The number of parameters in this fine-tuning portion is relatively small (e.g., the number of parameters in this portion is below or equal to a threshold, or the ratio of the number of parameters in this portion to the total number of parameters in the model is below or equal to a threshold) and / or the number of adjustment rounds is relatively small (e.g., the number of adjustment rounds is less than or equal to a threshold). Optionally, model fine-tuning can be implemented online. For example, during model fine-tuning, the communication device can fine-tune, adjust, or update some parameters of the model online.

[0283] As an example, the second condition is associated with the inference state information of the first model. Accordingly, the second condition is satisfied if at least one of the following is true: the performance of the first model is lower than or equal to a fourth threshold, the number of inferences of the first model is greater than or equal to a fifth threshold, or the inference duration of the first model is greater than or equal to a sixth threshold.

[0284] Optionally, the fourth threshold may be less than or equal to the first threshold.

[0285] Optionally, the fifth threshold may be greater than or equal to the second threshold.

[0286] Optionally, the sixth threshold is greater than or equal to the third threshold.

[0287] In another possible implementation of Example 2, if the second condition is met and the fourth condition is not met, the second model is obtained by fine-tuning the first model. That is, the first communication device can update the first model to the second model by fine-tuning the model to achieve the model update.

[0288] Optionally, the fine-tuned second model is adapted to the inference configuration information so that the inference process of the second model can obtain inference results that conform to the inference configuration information, thereby improving inference performance.

[0289] Optionally, in implementing Example 2, the method shown in Figure 3 also includes:

[0290] S304. The first communication device sends a fifth message, and correspondingly, the second communication device receives the fifth message. The fifth message is used to instruct the first model to obtain the second model through model fine-tuning. Therefore, the first communication device can also send a fifth message, enabling it to instruct or register the second model with the recipient of the fifth message, so that the recipient can subsequently schedule the second model obtained through model fine-tuning.

[0291] Optionally, before step S304, the method in Figure 3 further includes: the first communication device sending a first request message, the first request message being used to request first data, the first data being used for model fine-tuning; and the first communication device receiving the first data. Thus, when the first communication device determines that model fine-tuning of the first model is required, the first communication device can send the first request message, enabling the recipient of the first request message to send the first data based on the first request message, so that the first communication device can perform model fine-tuning of the first model based on the first data.

[0292] Example 3: The second model is determined through model training.

[0293] In one possible implementation of Example 3, if the third condition is met, the second model is obtained by training the first model. Therefore, if the third condition is met, the first communication device can update the first model to the second model through model training, thereby achieving model updating.

[0294] In this embodiment of the application, the communication device can support model training of one or more models. That is, after the communication device deploys one or more models, it can train the one or more models to obtain one or more trained models, and perform model inference based on the one or more trained models.

[0295] In the above process, the model training process may include: a communication device training a supported model to obtain a trained model.

[0296] For example, training a model using a communication device can be understood as the device training, adjusting, or updating some or all of the model's parameters. This involves a large number of parameters (e.g., the number of parameters is higher than or equal to a threshold, or the ratio of the number of parameters to the total number of parameters in the model is higher than or equal to a threshold) and / or a large number of adjustment rounds (e.g., the number of adjustment rounds is greater than or equal to a threshold). Optionally, model training can be implemented offline. For example, during model training, the communication device can train, adjust, or update some or all of the model's parameters offline.

[0297] As an example, the third condition is associated with the inference state information of the first model. Accordingly, the third condition is satisfied if at least one of the following is true: the performance of the first model is lower than or equal to the seventh threshold, the number of inferences of the first model is greater than or equal to the eighth threshold, or the inference duration of the first model is greater than or equal to the ninth threshold.

[0298] Optionally, the seventh threshold may be less than or equal to the fourth threshold.

[0299] Optionally, the eighth threshold is greater than or equal to the fifth threshold.

[0300] Optionally, the ninth threshold is greater than or equal to the sixth threshold.

[0301] In another possible implementation of Example 3, if the third condition is met and the fourth condition is not met, the second model is obtained by training the first model. That is, the first communication device can update the first model to the second model through model training to achieve model update.

[0302] Optionally, the second model is adapted to the inference configuration information so that the inference process of the second model can obtain the inference results that conform to the inference configuration information, thereby improving inference performance.

[0303] Optionally, in implementing Example 3, the method shown in Figure 3 also includes:

[0304] S305. The first communication device sends a sixth message, and correspondingly, the second communication device receives the sixth message. The sixth message indicates that the first model has been trained to obtain the second model. Therefore, the first communication device can also send the sixth message, enabling it to instruct or register the second model with the recipient of the fourth message, so that the recipient can subsequently schedule the trained second model.

[0305] Optionally, before step S305, the method shown in FIG3 further includes: the first communication device sending a second request message, the second request message being used to request second data, the first data being used for model training; and the first communication device receiving the second data. Thus, when the first communication device determines that model training is to be performed on the first model, the first communication device can send the second request message, enabling the recipient of the second request message to send the second data based on the second request message, so that the first communication device can perform model training on the second model based on the second data.

[0306] As can be seen from the above implementation process, the solution involved in this application can be applied to a communication device to deploy a model, and the communication device can interact with another communication device to realize the process of model usage and model updating. More examples will be described below.

[0307] As shown in Figure 4a, the process of model usage and model updating includes the following steps.

[0308] Step 1. Capability Interaction.

[0309] For example, the model deployment side sends descriptions of the single-sided models it supports to the other side, including at least one of the following: supported models, supported model update methods, and other side inference configuration methods. Supported models include supported functions, supported models, or models under a specific supported function; supported model update methods may include model switching, model fine-tuning, and model training; other side inference configuration methods may include not providing inference configuration, providing auxiliary inference information, or specifying an inference model.

[0310] In the embodiments of this application, step 1 is an implementation example of step A above, and the implementation processes of these two steps can be implemented by mutual reference.

[0311] As an example, as shown in Figure 4b, taking the communication device deploying the model (i.e., the first communication device) as the terminal device and the network device on the other side as an example, the capability interaction process may involve the following steps. For example, the network device queries the User Equipment Capability Enquiry for the models supported by the terminal device, the supported model update methods, and the inference configuration methods on the other side.

[0312] For example, UECapabilityEnquiry can be an implementation example of the request information used to request the fifth piece of information mentioned above.

[0313] For example, add the fields shown in Table 2 below to UECapabilityEnquiry.

[0314] Table 2

[0315] In addition, terminal devices can report their supported models, supported model update methods, and peer inference configuration methods in the User Equipment Capability Information (UECapabilityInformation). For example, UECapabilityInformation can be an implementation example of the fifth piece of information mentioned above. The fields shown in Table 3 below can be added to UECapabilityInformation.

[0316] Table 3

[0317] As another example, as shown in Figure 4c, taking the communication device deploying the model (i.e., the first communication device) as the network device and the counterpart device as the terminal device as an example, the capability interaction process is shown in Figure 4c. The network device can send the supported models, supported model update methods, and counterpart inference configuration methods to the terminal device in RRCReconfiguration. For example, RRCReconfiguration can be an implementation example of the fifth information mentioned above. For example, the fields shown in Table 4 below can be added to RRCReconfiguration.

[0318] Table 4

[0319] Step 2. Inference Configuration. The model deployment side (such as terminal devices or network devices) obtains inference configuration information, including one or more of the following: scene information (e.g., indoor / outdoor, idle / busy time, etc.), model information (e.g., at least one of model ID, model performance, model complexity), or system information (e.g., antenna configuration).

[0320] Step 3. Model Inference. The model deployment side (such as terminal devices or network devices) selects a model based on the inference configuration information and completes the model inference.

[0321] Step 4. Performance Monitoring. Monitor performance information when using a single-sided model on either side. This performance information includes model performance (such as one or more of inference accuracy or inference latency) and / or system performance (such as one or more of throughput, packet loss rate, latency, or fairness), so that the model deployment side can obtain this performance information.

[0322] Step 5. Model Update. The model deployment side determines the model update conditions based on performance information (such as performance degradation, the counter for using the model, or the timer for using the model expiring, etc.) and initiates the model update when these conditions are met.

[0323] As can be seen from the above implementation examples one to three, there may be three types of model update methods involved in this application, including model switching, model fine-tuning and model training. The following will describe them in conjunction with the example shown in Figure 4d.

[0324] As shown in Figure 4d, the selection of model update method and model update process may include any one of the following ①, ② or ③.

[0325] ① Model switching.

[0326] A1. The second communication device sends inference configuration information to the first communication device.

[0327] A2. The first communication device performs model switching based on the inference configuration information. For example, if the first communication device has a model that is compatible with it, it directly performs model switching. In this embodiment, step A2 is an implementation example of step S302 above.

[0328] A3. The first communication device sends a model switching instruction (or model update instruction) to the second communication device. Step A3 is optional. In this embodiment, step A3 is an example of an implementation of step S303 above.

[0329] For example, in step A3, the first communication device can send a model switching instruction (or model update instruction) to the second communication device through downlink control information (DCI), uplink control information (UCI), or user equipment assistance information (UAI) to indicate to the other side that the model has been updated.

[0330] ② Model fine-tuning.

[0331] B1. The first communication device performs model fine-tuning. For example, if the first communication device determines that the model inference performance meets the fine-tuning conditions (such as model performance loss being less than a threshold, the counter using the model being less than a threshold, or the timer using the model being less than a threshold), then model fine-tuning is initiated (optionally, the conditions for initiating fine-tuning may also include that the first communication device does not have a model adapted to the inference configuration information). In this embodiment, step B1 is an implementation example of step S302 above.

[0332] B2. The first communication device sends a model fine-tuning instruction (or model update instruction) to the second communication device. Step B2 is optional. In this embodiment, step B2 is an example of an implementation of step S304 above.

[0333] For example, in step B2, the first communication device can send a model switching instruction (or model fine-tuning instruction) to the second communication device via DCI, UCI or UAI to indicate to the other side that the model has been updated.

[0334] Optionally, during the model fine-tuning process, the first communication device sends a fine-tuning data acquisition instruction to the second communication device; correspondingly, the second communication device sends fine-tuning data to the first communication device, which is used for model fine-tuning.

[0335] ③ Model training.

[0336] C1. The first communication device performs model training. For example, if the first communication device determines that the model inference performance meets the fine-tuning conditions (such as the model performance loss being greater than a threshold, the counter using the model being greater than a threshold, or the timer using the model being greater than a threshold), then model fine-tuning is initiated (optionally, the conditions for initiating fine-tuning may also include that the first communication device does not have a model adapted to the inference configuration information). In this embodiment, step C1 is an implementation example of step S302 above.

[0337] C2. The first communication device sends a model training instruction (or model update instruction) to the second communication device. Step C2 is optional. In this embodiment, step C2 is an example of an implementation of step S305 above.

[0338] For example, in step C2, the first communication device can send a model training instruction (or model update instruction) to the second communication device via DCI, UCI, or UAI to indicate to the other side that the model has been updated.

[0339] Optionally, during the model training process, the first communication device sends a training data acquisition instruction to the second communication device; correspondingly, the second communication device sends training data to the first communication device, which is used for model training.

[0340] C3. The first communication device sends a rollback instruction to the second communication device. Step C3 is optional. For example, since model training requires a certain amount of time, during which the first communication device may not be able to perform intelligent processing (e.g., AI model processing, ML model processing, etc.) using the model, in step C3, the first communication device may instruct the other side to roll back to non-AI processing (e.g., non-AI model processing, non-ML model processing, etc.).

[0341] To facilitate understanding of the above solution, more examples will be used to describe it below.

[0342] Please refer to the example shown in Figure 4e, which uses a network device as the first communication device for deploying the model and a terminal device as the second communication device. In this example, the model deployed by the network device is a radio frequency (RF) map model, and the model update method is model fine-tuning.

[0343] Step 1. Capability Interaction. For example, the network device sends the supported models, supported model update methods, and inference configuration methods to the terminal device via RRCReconfiguration. Among the supported models is the RF map model; in the supported model update methods, the model-finetuning field is TRUE, meaning the network device supports model fine-tuning; in the inference configuration method, the additional-info field is TRUE; and the additional-info-Content field contains the terminal device's location information.

[0344] Step 2. Inference Configuration. For example, the network device obtains inference configuration information, which includes location information reported by the terminal device, and may also include one or more of the following: scene information, environment information, or system information.

[0345] Step 3. Model Inference. For example, based on the inference configuration information, the network device selects a suitable RF map model and performs model inference.

[0346] Step 4. Network devices and terminal devices begin transmission based on the RF map model.

[0347] Step 5. Performance monitoring. For example, network devices and terminal devices perform performance monitoring during RF map-based transmission.

[0348] Step 6. Enable model update. For example, when the detected performance degradation or inference count reaches a preset value A, and the network device has no other available RF map model that matches the inference configuration information, and the performance degradation is less than threshold B or the inference count is less than threshold C; therefore, the network device decides to fine-tune the model.

[0349] Optionally, the process shown in Figure 4e may also include steps 7 to 9 below.

[0350] Step 7. The network device sends a model fine-tuning instruction to the terminal device. For example, the network device sends the model fine-tuning instruction to the terminal device via RRC or DCI.

[0351] Step 8. The network device sends a fine-tuning data acquisition instruction to the terminal device.

[0352] Step 9. The terminal device sends fine-tuning data to the network device so that the network device can fine-tune the RF map model.

[0353] Step 10. The network device performs model fine-tuning on the RF map model.

[0354] Please refer to the example shown in Figure 4f, which uses a network device as the first communication device for deploying the model and a terminal device as the second communication device. In this example, the model deployed by the network device is an RF map model, and the model update method is model training.

[0355] Step 1. Capability Interaction. For example, network devices can send the models they support, supported model update methods, and inference configuration methods to terminal devices via RRCReconfiguration. Among the supported models is the RF map model; in the supported model update methods, the model-training field is TRUE, meaning the network device supports model training; in the inference configuration method, the additional-info field is TRUE; and the additional-info-Content field contains UE location information.

[0356] Step 2. Inference Configuration. The network device obtains inference configuration information, which includes location information reported by the terminal device, and may also include at least one of scene information, environment information, or system information.

[0357] Step 3. Model Inference. For example, based on the inference configuration information, the network device selects a suitable RF map model and performs model inference.

[0358] Step 4. Network devices and terminal devices begin transmission based on the RF map model.

[0359] Step 5. Performance monitoring. For example, network devices and terminal devices perform performance monitoring during RF map-based transmission.

[0360] Step 6. Enable model update. For example, when the detected performance degradation or the number of inferences reaches a preset value A, and the network device has no other available RF map model that matches the inference configuration information, and the performance degradation is greater than or equal to threshold B or the number of inferences is greater than or equal to threshold C, the network device decides to train the model.

[0361] Optionally, the process shown in Figure 4f may also include steps 7 to 9 below.

[0362] Step 7. The network device sends a model training instruction to the terminal device. For example, the network device sends a model training instruction to the terminal device via RRC or DCI.

[0363] Step 8. The network device sends a training data collection instruction to the terminal device.

[0364] Step 9. The terminal device sends training data to the network device so that the network device can fine-tune the RF map model.

[0365] Step 10. The network device trains the RF map model. Optionally, since model training takes a certain amount of time, the network device may not be able to perform intelligent processing (such as AI model processing, ML model processing, etc.) using the RF map model during this time. Therefore, the network device may instruct the terminal device to fall back to non-AI processing (such as non-AI model processing, non-ML model processing, etc.).

[0366] Please refer to the example shown in Figure 4g, taking the first communication device deploying the model as the terminal device and the second communication device as the network device. In this example, the model deployed on the terminal device is a channel estimation model, and the model update method is model fine-tuning.

[0367] Step 1. Capability Interaction. For example, the network device queries the UECapabilityEnquiry to find the models supported by the terminal device, the supported model update methods, and the peer inference configuration methods. Subsequently, the terminal device reports its supported models, supported model update methods, and peer inference configuration methods in the UECapabilityInformation. For example, the models supported by the terminal device include channel estimation models; in the supported model update methods, the model-finetuning field is TRUE, meaning the terminal device supports model fine-tuning; in the inference configuration methods, the additional-info field and the specified-model field are both TRUE; the additional-info-Content field contains system configuration information, and the specified-model-ID field contains the model ID.

[0368] Step 2. Inference Configuration. For example, the network device sends inference configuration information to the terminal device, which includes system configuration information and model ID issued by the network device, and may also include one or more of the following: scene information and environment information.

[0369] Step 3. Model Inference. For example, based on the inference configuration information, the terminal device selects a suitable channel estimation model and performs model inference.

[0370] Step 4. Network devices and terminal devices begin transmission based on the channel estimation model.

[0371] Step 5. Performance monitoring. For example, network devices and terminal devices perform performance monitoring during transmission based on channel estimation models.

[0372] Step 6. Enable model update. For example, when the detected performance degradation or inference count reaches a preset value A, and the terminal device has no other available channel estimation model that matches the inference configuration information, and the performance degradation is less than threshold B or the inference count is less than threshold C, the terminal device decides to fine-tune the model.

[0373] Optionally, the process shown in Figure 4g may also include step 7 below.

[0374] Step 7. The terminal device sends a model fine-tuning instruction to the network device. For example, the terminal device reports the model fine-tuning instruction to the network device via RRC, UCI, or UAI.

[0375] Step 8. The terminal device fine-tunes the channel estimation model.

[0376] Please refer to the example shown in Figure 4h, taking the first communication device deploying the model as the terminal device and the second communication device as the network device. In this example, the model deployed on the terminal device is a channel estimation model, and the model update method is model training.

[0377] Step 1. Capability Interaction. The network device queries the UECapabilityEnquiry for the models supported by the terminal device, the supported model update methods, and the peer inference configuration methods. Subsequently, the terminal device reports its supported models, supported model update methods, and peer inference configuration methods in the UECapabilityInformation. Among the models supported by the terminal device are channel estimation models. In the model update methods supported by the terminal device, the value of the model-training field is TRUE, meaning that the terminal device supports model training. In the inference configuration methods, the value of the additional-info field and the specified-model field are both TRUE. The additional-info-Content field contains system configuration information, and the specified-model-ID field contains the model ID.

[0378] Step 2: Inference Configuration. For example, the network device sends inference configuration information to the terminal device, which includes system configuration information and model ID issued by the network device, and may also include one or more of the following: scene information and environment information.

[0379] Step 3. Model Inference. For example, based on the inference configuration information, the terminal device selects a suitable channel estimation model and performs model inference.

[0380] Step 4. Network devices and terminal devices begin transmission based on the channel estimation model.

[0381] Step 5. Performance monitoring. For example, network devices and terminal devices perform performance monitoring during transmission based on channel estimation models.

[0382] Step 6. Enable model update. For example, when the detected performance degradation or the number of inference attempts reaches a preset value A, and the terminal device has no other available channel estimation model that matches the inference configuration information, and the performance degradation is greater than or equal to threshold B or the number of inference attempts is greater than or equal to threshold C, the terminal device decides to perform model training.

[0383] Optionally, the process shown in Figure 4h also includes step 7 below.

[0384] Step 7. The terminal device sends a model training instruction to the network device. For example, the terminal device reports the model training instruction to the network device via RRC, UCI, or UAI.

[0385] Step 8. The terminal device trains the channel estimation model. Optionally, since model training takes a certain amount of time, the terminal device may not be able to perform intelligent processing (such as AI model processing, ML model processing, etc.) through the channel estimation model during this time. Therefore, the terminal device reports a backoff request through RRC, UCI, or UAI, and the network device issues a backoff instruction through RRC or DCI.

[0386] Referring to Figure 5, this application embodiment provides a communication device 500. This communication device 500 can implement the functions of the first communication device (or second communication device) in the above method embodiments, and therefore can also achieve the beneficial effects of the above method embodiments. In this application embodiment, the communication device 500 can be the first communication device (or the second communication device), or it can be an integrated circuit or component inside the first communication device (or the second communication device), such as a chip, baseband chip, modem chip, SoC chip (e.g., an SoC chip containing a modem core), SIP chip, communication module, chip system, processor, etc.

[0387] Optionally, the transceiver unit 502 may include a transmitting unit and a receiving unit, which are used to perform transmitting and receiving respectively.

[0388] In one possible implementation, when the device 500 is used to execute the method performed by the first communication device in the preceding embodiments, the device 500 includes a processing unit 501; the processing unit 501 is used to acquire first information, the first information being used to instruct updating (or processing, optimizing, iterating) the first model to a second model; wherein, the first information is related to the inference state information of the first model; the processing unit 501 is also used to determine a second model based on the inference state information of the first model, the second model being different from the first model.

[0389] In one possible implementation, when the device 500 is used to execute the method performed by the second communication device in the preceding embodiments, the device 500 includes a processing unit 501 and a transceiver unit 502; the transceiver unit 502 is used to receive AI capability information of the first communication device (e.g., the AI ​​capability information of the first communication device is included in the preceding second or third information); wherein, the AI ​​capability information indicates a supported inference configuration method, and the supported inference configuration method indicates whether it is supported for other devices to specify inference configuration information; if the processing unit 501 determines that the supported inference configuration method indicates support for other devices to specify inference configuration information, the transceiver unit 502 is further used to send the inference configuration information, which is used to determine a second model; wherein, the second model is obtained by updating the first model.

[0390] In one possible implementation, when the device 500 is used to execute the method performed by the second communication device in the preceding embodiments, the device 500 includes a transceiver unit 502; the processing unit 501 is used to determine inference configuration information; the transceiver unit 502 is used to send the inference configuration information; wherein the inference configuration information is used to determine the second model; wherein the second model is obtained by updating the first model.

[0391] In one possible implementation, when the device 500 is used to execute the method performed by the second communication device in the preceding embodiments, the device 500 includes a processing unit 501 and a transceiver unit 502; the processing unit 501 is used to determine first information, which is used to instruct updating the first model to the second model; wherein, the first information is related to the inference state information of the first model, which includes at least one of the performance of the first model, the number of times model inference is performed using the first model, or the duration of model inference performed using the first model; the transceiver unit 502 is used to send the first information.

[0392] In one possible design, when the communication device 500 is a terminal device or a communication module within a terminal, the functionality of the processing unit 501 can be implemented by one or more processors. Specifically, the processor may include a modem chip, a SoC chip (such as a SoC chip containing a modem core), or a SIP chip. The functionality of the transceiver unit 502 can be implemented by transceiver circuitry.

[0393] In one possible design, when the communication device 500 is a circuit or chip in a terminal responsible for communication functions, such as a modem chip, a SoC chip, or a SoC chip or SIP chip containing a modem core, the function of the processing unit 501 can be implemented by a circuit system in the aforementioned chip that includes one or more processors or processor cores. The function of the transceiver unit 502 can be implemented by the interface circuitry or data transceiver circuitry on the aforementioned chip.

[0394] Optionally, the details of the information execution process of the unit of the communication device 500 can be found in the description of the method embodiment shown above in this application, and will not be repeated here.

[0395] Please refer to Figure 6, which is another schematic structural diagram of the communication device 600 provided in this application. The communication device 600 includes a logic circuit 601 and an input / output interface 602. The communication device 600 can be a chip or an integrated circuit.

[0396] In Figure 5, the transceiver unit 502 can be a communication interface, which can be the input / output interface 602 in Figure 6, and the input / output interface 602 can include an input interface and an output interface. Alternatively, the communication interface can also be a transceiver circuit, which can include an input interface circuit and an output interface circuit.

[0397] In one possible implementation, when the device 600 is used to execute the method performed by the first communication device in the preceding embodiments, the logic circuit 601 is used to acquire first information, which is used to instruct the first model to be updated (or processed, optimized, iterated) to a second model; wherein, the first information is related to the inference state information of the first model; the logic circuit 601 is also used to determine a second model based on the inference state information of the first model, the second model being different from the first model.

[0398] In one possible implementation, when the device 600 is used to execute the method performed by the second communication device in the preceding embodiments, the input / output interface 602 is used to receive AI capability information of the first communication device (e.g., the AI ​​capability information of the first communication device is included in the preceding second or third information); wherein, the AI ​​capability information indicates a supported inference configuration method, and the supported inference configuration method indicates whether it supports the specification of inference configuration information by other devices; when the logic circuit 601 determines that the supported inference configuration method indicates support for the specification of inference configuration information by other devices, the input / output interface 602 is also used to send the inference configuration information, which is used to determine a second model; wherein, the second model is obtained by updating the first model.

[0399] In one possible implementation, when the device 600 is used to execute the method performed by the second communication device in the preceding embodiments, the logic circuit 601 is used to determine inference configuration information; the input / output interface 602 is used to send the inference configuration information; wherein the inference configuration information is used to determine the second model; wherein the second model is obtained by updating the first model.

[0400] In one possible implementation, when the device 600 is used to execute the method performed by the second communication device in the preceding embodiments, the logic circuit 601 is used to determine first information, which is used to instruct updating the first model to the second model; wherein, the first information is related to the inference state information of the first model, which includes at least one of the performance of the first model, the number of times model inference is performed using the first model, or the duration of model inference performed using the first model; the input / output interface 602 is used to send the first information.

[0401] The logic circuit 601 and the input / output interface 602 can also perform other steps performed by the first or second communication device in any embodiment and achieve corresponding beneficial effects, which will not be elaborated here.

[0402] In one possible implementation, the processing unit 501 shown in FIG5 can be the logic circuit 601 in FIG6.

[0403] Optionally, the logic circuit 601 can be a processing device, the functions of which can be partially or entirely implemented in software.

[0404] Optionally, the processing apparatus may include a memory and a processor, wherein the memory is used to store a computer program, and the processor reads and executes the computer program stored in the memory to perform the corresponding processing and / or steps in any of the method embodiments.

[0405] Optionally, the processing device may consist of only a processor. A memory for storing computer programs is located outside the processing device, and the processor is connected to the memory via circuitry / wires to read and execute the computer programs stored in the memory. The memory and processor may be integrated together or physically independent of each other.

[0406] Optionally, the processing device may be one or more chips, or one or more integrated circuits. For example, the processing device may be one or more field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), system-on-chips (SoCs), central processing units (CPUs), network processors (NPs), digital signal processors (DSPs), microcontroller units (MCUs), programmable logic devices (PLDs), or other integrated chips, or any combination of the above chips or processors.

[0407] Please refer to Figure 7, which shows the communication device 700 involved in the above embodiments provided in the embodiments of this application. Specifically, the communication device 700 can be the communication device as a terminal device in the above embodiments. The example shown in Figure 7 is that the terminal device is implemented through the terminal device (or the components in the terminal device).

[0408] The present invention is a possible logical structure diagram of the communication device 700, which may include, but is not limited to, at least one processor 701 and a communication port 702.

[0409] In Figure 5, the transceiver unit 502 can be a communication interface, which can be the communication port 702 in Figure 7. The communication port 702 can include an input interface and an output interface. Alternatively, the communication port 702 can also be a transceiver circuit, which can include an input interface circuit and an output interface circuit.

[0410] Further optionally, the device may also include at least one of a memory 703 and a bus 704. In the embodiments of this application, the at least one processor 701 is used to control the operation of the communication device 700.

[0411] Furthermore, the processor 701 can be a central processing unit, a general-purpose processor, a digital signal processor, an application-specific integrated circuit, a field-programmable gate array, or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. The processor can also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a digital signal processor and a microprocessor, etc. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0412] Optionally, the communication device 700 shown in FIG7 can be used to implement the steps implemented by the terminal device in the aforementioned method embodiments and achieve the corresponding technical effects of the terminal device. The specific implementation of the communication device shown in FIG7 can be referred to the description in the aforementioned method embodiments, and will not be repeated here.

[0413] Please refer to Figure 8, which is a schematic diagram of the structure of the communication device 800 involved in the above embodiments provided in the embodiments of this application. Specifically, the communication device 800 can be a communication device as a network device in the above embodiments. The example shown in Figure 8 is that the network device is implemented through a network device (or a component in the network device). The structure of the communication device can refer to the structure shown in Figure 8.

[0414] The communication device 800 includes at least one processor 811 and at least one network interface 814. Optionally, the communication device further includes at least one memory 812, at least one transceiver 813, and one or more antennas 815. The processor 811, memory 812, transceiver 813, and network interface 814 are connected, for example, via a bus. In this embodiment, the connection may include various interfaces, transmission lines, or buses, etc., and this embodiment is not limited thereto. The antenna 815 is connected to the transceiver 813. The network interface 814 enables the communication device to communicate with other communication devices through a communication link. For example, the network interface 814 may include a network interface between the communication device and core network equipment, such as an S1 interface, or a network interface between the communication device and other communication devices (e.g., other network devices or core network equipment), such as an X2 or Xn interface.

[0415] In Figure 5, the transceiver unit 502 can be a communication interface, which can be the network interface 814 in Figure 8. The network interface 814 can include an input interface and an output interface. Alternatively, the network interface 814 can also be a transceiver circuit, which can include an input interface circuit and an output interface circuit.

[0416] The processor 811 is primarily used to process communication protocols and communication data, control the entire communication device, execute software programs, and process data from these programs, for example, to support the actions described in the embodiments of the communication device. The communication device may include a baseband processor and a central processing unit (CPU). The baseband processor is primarily used to process communication protocols and communication data, while the CPU is primarily used to control the entire terminal device, execute software programs, and process data from these programs. The processor 811 in Figure 8 can integrate the functions of both a baseband processor and a CPU. Those skilled in the art will understand that the baseband processor and CPU can also be independent processors interconnected via technologies such as buses. Those skilled in the art will understand that a terminal device may include multiple baseband processors to adapt to different network standards, and multiple CPUs to enhance its processing capabilities. The various components of the terminal device can be connected via various buses. The baseband processor can also be described as a baseband processing circuit or a baseband processing chip. The CPU can also be described as a central processing circuit or a central processing chip. The function of processing communication protocols and communication data can be built into the processor or stored in memory as a software program, which is then executed by the processor to implement the baseband processing function.

[0417] The memory is primarily used to store software programs and data. The memory 812 can exist independently or be connected to the processor 811. Optionally, the memory 812 can be integrated with the processor 811, for example, integrated into a single chip. The memory 812 can store program code that executes the technical solutions of the embodiments of this application, and its execution is controlled by the processor 811. The various types of computer program code being executed can also be considered as drivers for the processor 811.

[0418] Figure 8 shows only one memory and one processor. In the terminal device of this application embodiment, there may be multiple processors and multiple memories. Memory may also be called storage medium or storage device, etc. Memory can be a storage element on the same chip as the processor, i.e., an on-chip storage element, or it can be a separate storage element; this application embodiment does not limit this.

[0419] Transceiver 813 can be used to support the reception or transmission of radio frequency (RF) signals between a communication device and a terminal. Transceiver 813 can be connected to antenna 815. Transceiver 813 includes a transmitter Tx and a receiver Rx. Specifically, one or more antennas 815 can receive RF signals. The receiver Rx of transceiver 813 receives the RF signals from the antennas, converts the RF signals into digital baseband signals or digital intermediate frequency (IF) signals, and provides the digital baseband signals or IF signals to processor 811 so that processor 811 can perform further processing on the digital baseband signals or IF signals, such as demodulation and decoding. Furthermore, the transmitter Tx in transceiver 813 is also used to receive modulated digital baseband signals or IF signals from processor 811, convert the modulated digital baseband signals or IF signals into RF signals, and transmit the RF signals through one or more antennas 815. Specifically, the receiver Rx can selectively perform one or more stages of downmixing and analog-to-digital conversion on the radio frequency signal to obtain a digital baseband signal or a digital intermediate frequency (IF) signal. The order of these downmixing and IF conversion processes is adjustable. The transmitter Tx can selectively perform one or more stages of upmixing and digital-to-analog conversion on the modulated digital baseband signal or digital IF signal to obtain a radio frequency signal. The order of these upmixing and IF conversion processes is also adjustable. The digital baseband signal and the digital IF signal can be collectively referred to as digital signals.

[0420] The transceiver 813 can also be called a transceiver unit, transceiver, transceiver device, etc. Optionally, the device in the transceiver unit that performs the receiving function can be regarded as the receiving unit, and the device in the transceiver unit that performs the transmitting function can be regarded as the transmitting unit. That is, the transceiver unit includes a receiving unit and a transmitting unit. The receiving unit can also be called a receiver, input port, receiving circuit, etc., and the transmitting unit can be called a transmitter, transmitter, or transmitting circuit, etc.

[0421] Optionally, the communication device 800 shown in FIG8 can be used to implement the steps implemented by the network device in the aforementioned method embodiments and achieve the corresponding technical effects of the network device. The specific implementation of the communication device 800 shown in FIG8 can be referred to the description in the aforementioned method embodiments, and will not be repeated here.

[0422] Please refer to Figure 9, which is a schematic diagram of the structure of the communication device involved in the above embodiments provided in the embodiments of this application.

[0423] The communication device 900 includes, for example, modules, units, elements, circuits, or interfaces, appropriately configured together to execute the technical solutions provided in this application. The communication device 900 may be the terminal device or network device described above, or a component (e.g., a chip) within these devices, used to implement the methods described in the following method embodiments. The communication device 900 includes one or more processors 901. The processor 901 may be a general-purpose processor or a dedicated processor, for example, a baseband processor or a central processing unit. The baseband processor can be used to process communication protocols and communication data, while the central processing unit can be used to control the communication device (e.g., a RAN node, terminal, or chip), execute software programs, and process data from the software programs.

[0424] Optionally, in one design, processor 901 may include program 903 (sometimes also referred to as code or instructions), which may be executed on processor 901 to cause communication device 900 to perform the methods described in the embodiments below. In yet another possible design, communication device 900 includes circuitry (not shown in FIG9).

[0425] Optionally, the communication device 900 may include one or more memories 902 storing a program 904 (sometimes referred to as code or instructions), which can be run on the processor 901 to cause the communication device 900 to perform the methods described in the above method embodiments.

[0426] Optionally, the processor 901 and / or memory 902 may include AI modules 907 and 908, which are used to implement AI-related functions. The AI ​​modules can be implemented through software, hardware, or a combination of both. For example, the AI ​​module may include a radio intelligence control (RIC) module. For example, the AI ​​module may be a near real-time RIC or a non-real-time RIC.

[0427] Optionally, the processor 901 and / or memory 902 may also store data. The processor and memory may be configured separately or integrated together.

[0428] Optionally, the communication device 900 may further include a transceiver 905 and / or an antenna 906. The processor 901, sometimes referred to as a processing unit, controls the communication device (e.g., a RAN node or terminal). The transceiver 905, sometimes referred to as a transceiver unit, transceiver, transceiver circuit, or transceiver, is used to implement the transmission and reception functions of the communication device via the antenna 906.

[0429] In Figure 5, the processing unit 501 can be a processor 901. The transceiver unit 502 shown in Figure 5 can be a communication interface, which can be the transceiver 905 in Figure 9. The transceiver 905 can include an input interface and an output interface. Alternatively, the transceiver 905 can also be a transceiver circuit, which can include an input interface circuit and an output interface circuit.

[0430] This application also provides a computer-readable storage medium for storing one or more computer-executable instructions. When the computer-executable instructions are executed by a computer, the computer performs the method described in the possible implementations of the first or second communication device in the foregoing embodiments.

[0431] This application also provides a computer program product (or computer program) that, when executed by a computer, allows the computer to execute the method described above for the possible implementation of the first or second communication device.

[0432] This application also provides a chip system including at least one processor for supporting a communication device in implementing the functions involved in the possible implementations of the communication device described above. Optionally, the chip system further includes an interface circuit that provides program instructions and / or data to the at least one processor. In one possible design, the chip system may also include a memory for storing the program instructions and data necessary for the communication device. The chip system may be composed of chips or may include chips and other discrete devices, wherein the communication device may specifically be the first communication device or the second communication device in the aforementioned method embodiments.

[0433] This application also provides a communication system, which includes the first communication device in any of the above embodiments.

[0434] Optionally, the communication system may also include a second communication device.

[0435] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces, or indirect coupling or communication connection between devices or units, and may be electrical, mechanical, or other forms. Whether a function is implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0436] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected as needed to achieve the purpose of this embodiment.

[0437] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

Claims

1. A communication method characterized by comprising: include: Obtain first information, which is used to indicate that the first model is updated to the second model; wherein the first information is related to the inference state information of the first model, and the inference state information of the first model includes at least one of the following: the performance of the first model, the number of times the first model is used for model inference, or the duration of model inference using the first model. Based on the first information, the first model is updated to the second model.

2. The method of claim 1, wherein, The acquisition of the first information includes: Receive the first information.

3. The method of claim 2, wherein, The method further includes: Send a second message, the second message indicating the AI ​​capability information of the first communication device; wherein the first message is determined based on the inference state information of the first model and the AI ​​capability information of the first communication device.

4. The method of claim 3, wherein, The AI ​​capability information indicates at least one of the following: One or more supported models, wherein the one or more supported models include at least the first model; Supported model update methods include at least one of model switching, model fine-tuning, or model training; or Supported inference configuration methods, wherein the supported inference configuration methods indicate whether it is supported for other devices to specify inference configuration information, wherein the inference configuration information includes inference assistance information, or wherein the inference configuration information includes at least one of the following: model identifier.

5. The method of claim 4, wherein, The AI ​​capability information indicates supported inference configuration methods, and the supported inference configuration methods indicate support for specifying inference configuration information by other devices; the method further includes: The inference configuration information is received and used to determine the second model.

6. The method of claim 1, wherein, The acquisition of the first information includes: Obtain the inference state information of the first model; The first information is determined based on the reasoning state information of the first model.

7. The method of claim 6, wherein, The method further includes: Send a third message, the third message indicating the AI ​​capability information of the first communication device; wherein, the AI ​​capability information indicates the supported inference configuration method, and the supported inference configuration method indicates that inference configuration information can be specified by other devices; The method further includes: The inference configuration information is received and used to determine the second model.

8. The method according to any one of claims 1 to 7, characterized in that, The second model satisfies any of the following: If the first condition is met, the second model is one of one or more models supported by the first communication device; If the second condition is met, the second model is obtained by fine-tuning the first model. or If the third condition is met, the second model is obtained by training the model based on the first model.

9. The method according to any one of claims 1 to 7, characterized in that, The step of updating the first model to the second model based on the first information includes: Based on the first information and the inference configuration information, the first model is updated to the second model.

10. The method of claim 9, wherein, The second model satisfies any of the following: If the first condition is satisfied and the fourth condition is satisfied, the second model is one of one or more models supported by the first communication device. If the second condition is met and the fourth condition is not met, the second model is obtained by fine-tuning the first model. or If the third condition is met and the fourth condition is not met, the second model is obtained by training the model based on the first model. The fourth condition being satisfied includes: one or more models supported by the first communication device contain a model that is compatible with the inference configuration information; the fourth condition not being satisfied includes: one or more models supported by the first communication device do not contain a model that is compatible with the inference configuration information.

11. The method according to claim 8 or 10, characterized in that, Meet at least one of the following: The first condition includes at least one of the following: the performance of the first model is lower than or equal to a first threshold, the number of inferences of the first model is greater than or equal to a second threshold, or the inference time of the first model is greater than or equal to a third threshold. The second condition includes at least one of the following: the performance of the first model is lower than or equal to the fourth threshold; the number of inferences of the first model is greater than or equal to the fifth threshold; or the inference time of the first model is greater than or equal to the sixth threshold; or The fourth condition includes at least one of the following: the performance of the first model is lower than or equal to the seventh threshold, the number of inferences of the first model is greater than or equal to the eighth threshold, or the inference time of the first model is greater than or equal to the ninth threshold.

12. The method of any one of claims 4, 5, 7, 9, or 10, wherein, The inference configuration information includes one or more model identifiers and / or inference assistance information.

13. A method of communication, comprising: include: Receive AI capability information from a first communication device; wherein the AI ​​capability information indicates supported inference configuration methods, and the supported inference configuration methods indicate whether inference configuration information can be specified by other devices; When the supported inference configuration method indicates support for specifying inference configuration information by other devices, the inference configuration information is sent, and the inference configuration information is used to determine the second model; wherein the second model is obtained by updating the first model.

14. The method of claim 13, wherein, The second model is obtained by updating the first model, including: The second model is obtained by updating the first model based on the first information; wherein the first information is related to the inference state information of the first model, and the inference state information of the first model includes at least one of the following: the performance of the first model, the number of times the first model is used for model inference, or the duration of model inference using the first model.

15. The method of claim 14, wherein, The method further includes: Send the first message.

16. The method according to any one of claims 13 to 15, characterized in that, The AI ​​capability information indicates at least one of the following: One or more supported models, wherein the supported one or more models include at least the first model; or Supported model update methods include at least one of model switching, model fine-tuning, or model training.

17. The method according to any one of claims 13 to 16, characterized in that, The second model satisfies any of the following: If the first condition is met and the fourth condition is met, the second model is one of one or more models supported by the first communication device. If the second condition is met and the fourth condition is not met, the second model is obtained by fine-tuning the first model. or the second model is obtained based on the first model when the third condition is met and the fourth condition is not met; wherein the fourth condition is met includes that there is a model in the one or more models supported by the first communication device that is adapted to the inference configuration information; and the fourth condition is not met includes that there is no model in the one or more models supported by the first communication device that is adapted to the inference configuration information.

18. The method of claim 17, wherein, At least one of the following is met: the first condition includes at least one of that the performance of the first model is less than or equal to a first threshold, the number of inferences of the first model is greater than or equal to a second threshold, or the inference time length of the first model is greater than or equal to a third threshold; the second condition includes at least one of that the performance of the first model is less than or equal to a fourth threshold, the number of inferences of the first model is greater than or equal to a fifth threshold, or the inference time length of the first model is greater than or equal to a sixth threshold; or the fourth condition includes at least one of that the performance of the first model is less than or equal to a seventh threshold, the number of inferences of the first model is greater than or equal to an eighth threshold, or the inference time length of the first model is greater than or equal to a ninth threshold.

19. The method according to any one of claims 13 to 18, characterized in that, The inference configuration information includes one or more model identifiers and / or inference assistance information.

20. A communications device, characterized by A module for performing the method of any one of claims 1-19.

21. A communications device, characterized by At least one processor for performing the method of any one of claims 1-19.

22. The communication apparatus according to claim 21, wherein, The communication device is a chip or a chip system.

23. A computer-readable storage medium, characterized in that, The computer readable storage medium has stored therein a computer program or instructions which, when executed, implement the method of any one of claims 1-19.

24. A computer program product, characterised in that, A computer program or instructions which, when executed by a computer, implement the method of any one of claims 1-19.

Images