WO2025189861A1 - Communication method and related apparatus - Google Patents

Abstract

A communication method and a related apparatus. In the method, after a first communication apparatus acquires performance monitoring information of at least one AI model by means of first information, the first communication apparatus may perform noise perturbation processing on a model parameter of a first AI model, so as to obtain an updated model parameter of the first AI model. By using the method, the computing power of a communication node can be applied to the processing of an AI model, and the AI model can be updated by means of performance monitoring information of the AI model. In addition, by means of the process of performing noise perturbation processing on a parameter of the AI model, the problem of the plasticity of the AI model being reduced or even disappeared during a long-time continual learning process can be mitigated or avoided, the adaptability of continual learning can be improved, and the robustness of continual learning can also be improved.

Description

A communication method and related device

This application claims priority to the Chinese patent application filed with the State Intellectual Property Office on March 15, 2024, with application number 202410303028.6 and application name “A communication method and related device”, the entire contents of which are incorporated by reference into this application.

Technical Field

The present application relates to the field of communications, and in particular to a communication method and related devices.

Background Art

Wireless communication can be a transmission communication between two or more communication nodes without propagating through conductors or cables. The communication nodes generally include network devices and terminal devices.

Currently, in wireless communication systems, communication nodes generally possess both signal transceiver capabilities and computing capabilities. For example, network devices with computing capabilities primarily provide computing power to support signal transceiver capabilities (e.g., processing both sending and receiving signals), enabling communication between the network device and other communication nodes.

However, in communication networks, communication nodes may have excess computing power beyond just supporting the aforementioned communication tasks. Therefore, how to utilize this computing power is a pressing technical issue.

Summary of the Invention

The present application provides a communication method and related devices for improving the robustness of continuous learning.

In a first aspect, the present application provides a communication method, which is applied to a first communication device. The first communication device may be a communication device (such as a terminal device or a network device), or the first communication device may be a partial component in the communication device (such as a processor, a chip or a chip system, etc.), or the first communication device may also be a logic module or software that can realize all or part of the functions of the communication device. In this method, the first communication device obtains first information, and the first information is used to indicate performance monitoring information of at least one artificial intelligence (AI) model, and the at least one AI model is associated with a first AI model deployed on the first communication device; the first communication device performs noise perturbation processing on the model parameters of the first AI model based on the first information to obtain the updated model parameters of the first AI model.

Based on the above solution, after the first communication device obtains performance monitoring information of at least one AI model through the first information, the first communication device can perform noise perturbation processing on the model parameters of the first AI model to obtain updated model parameters of the first AI model. In this way, the computing power of the communication node can be applied to AI model processing, and the AI model can be updated based on the performance monitoring information of the AI model.

In addition, the process of the first communication device performing the AI model parameter update is triggered based on the performance monitoring information of at least one AI model, and during the AI model parameter update process, the first communication device obtains the updated AI model parameters through noise perturbation processing. In this way, the solution can be applied to the scenario of continuous learning, and by performing noise perturbation processing on the AI model parameters, the problem of reduced or even disappearance of the plasticity of the AI model during long-term continuous learning can be slowed down or avoided, which can improve the adaptability of continuous learning and enhance the robustness of continuous learning.

In this application, AI model can be replaced by other terms, such as neural network, neural network model, AI neural network model, machine learning model, or AI processing model.

In this application, noise disturbance processing can be replaced by other terms, such as noise addition processing, disturbance processing, noise disturbance processing, noise interference processing, or random disturbance processing.

In a possible implementation of the first aspect, the method further includes: the first communication device sending second information based on the first information, the second information being obtained by performing noise disturbance processing based on the first parameter, and the second information being used to update a second AI model; wherein the second AI model is used to be deployed in the second communication device.

Based on the above solution, the first communication device can also send second information obtained by performing noise perturbation processing based on the first parameter based on the first information, so that the recipient of the second information (such as the second communication device) can obtain the second information obtained by performing noise perturbation processing based on the first parameter, and update the second AI model based on the second information. In this way, the problem of reduced or even disappearance of the plasticity of the second AI model during long-term continuous learning can be alleviated or avoided, and the adaptability of continuous learning can be improved, and the robustness of continuous learning can be enhanced.

In a possible implementation manner of the first aspect, the at least one AI model includes the first AI model and/or the second AI model; wherein the first AI model is associated with the second AI model, and the second AI model is used to be deployed on the second communication device.

Based on the above solution, the first information is used to indicate performance monitoring information of at least one AI model, and when the performance monitoring information indicated by the first information satisfies a certain condition (e.g., the first condition and/or the second condition described below), the first information can be used to trigger noise perturbation processing of model parameters of the first AI model and/or the second AI model. Accordingly, the at least one AI model may include the first AI model and/or the second AI model. In this way, the first communication device can trigger noise perturbation processing of the AI model based on the performance of the associated AI model.

Optionally, the first AI model is associated with the second AI model, including any of the following:

The first AI model is a public model and the second AI model is a personalized model;

The first AI model is a personalized model and the second AI model is a public model;

The input of the first AI model includes the output of the second AI model;

The input of the second AI model includes the output of the first AI model;

The first AI model is a teacher model and the second AI model is a student model; or,

The first AI model is a student model and the second AI model is a teacher model.

In a possible implementation manner of the first aspect, the second information is obtained by processing a quantization result of the first parameter, or the second information is obtained by processing a transmission parameter of the first parameter.

Based on the above solution, the second information can be obtained by performing noise disturbance processing in a variety of ways to improve the flexibility of the solution implementation.

Optionally, the first parameter includes one or more of the following:

The updated model parameters of the second AI model, used to update the gradient information of the second AI model, used to update the intermediate feature information of the second AI model, used to update the intermediate gradient information of the second AI model, the inference result of the first AI model, the distillation loss information corresponding to the second AI model, and the task loss information of the second AI model.

In a possible implementation of the first aspect, the first communication device sending the second information based on the first information includes: when a first condition is met, the first communication device sending the second information based on the first information; wherein the first condition includes at least one of the following:

The training accuracy of the at least one AI model is lower than a first threshold;

The test accuracy of the at least one AI model is lower than a second threshold;

The system performance of the AI model system in which the at least one AI model resides is lower than a third threshold;

The system performance of the communication system where the communication device deploying the at least one AI model is located is lower than a fourth threshold;

A change in the neural network input-output distribution of the at least one AI model is greater than a fifth threshold;

The neural network weight distribution of the at least one AI model is greater than a sixth threshold;

The change in the neural network weight of the at least one AI model is lower than a seventh threshold.

Based on the above solution, when the first condition is met, the first communication device can determine that the current operation of the second AI model may be abnormal. To this end, the first communication device can send second information, so that the second communication device can obtain the second information obtained by noise perturbation processing based on the first parameter, and update the second AI model based on the second information. In this way, the problem of the second AI model's plasticity being reduced or even disappearing during long-term continuous learning can be alleviated or avoided, and the adaptability and robustness of continuous learning can be improved.

In a possible implementation manner of the first aspect, the method further includes: the first communication device receiving indication information indicating a second parameter, where the second parameter is used to process the first parameter to obtain the second information.

Based on the above scheme, the first communication device can also determine the second parameter through the received indication information, and process the first parameter based on the second parameter to obtain the second information, that is, the first communication device can determine the second parameter based on the indication of other communication devices (for example, the recipient of the second information), so that the recipient of the second information can obtain the corresponding second information through the specified second parameter.

It should be understood that the second parameter can be used to perform noise disturbance processing on the first parameter to obtain the second information. Accordingly, the second parameter can be understood as a noise disturbance parameter, a noise parameter, or a disturbance parameter, etc.

In a possible implementation manner of the first aspect, the method further includes: the first communication device sending request information for requesting the second parameter.

Based on the above solution, the first communication device can determine the second parameter through an interactive process of the request information and the above indication information.

Optionally, the second parameter is a preconfigured parameter.

In a possible implementation of the first aspect, the method further includes: the first communication device receiving indication information indicating a third parameter, where the third parameter is used to process the model parameters of the first AI model to obtain updated model parameters of the first AI model.

Based on the above scheme, the first communication device can also receive the third parameter determined by the received indication information, and perform noise disturbance processing on the model parameters of the first AI model based on the third parameter, that is, the first communication device can determine the third parameter based on the indication of other communication devices (such as the recipient of the second information), so that the first communication device can obtain the noise disturbance processing through the specified third parameter.

It should be understood that the third parameter can be used to process the model parameters of the first AI model to obtain the updated model parameters of the first AI model. Accordingly, the third parameter can be understood as a noise disturbance parameter, a noise parameter, or a disturbance parameter, etc.

In a possible implementation manner of the first aspect, the method further includes: the first communication device sending request information for requesting the third parameter.

Based on the above solution, the first communication device can determine the third parameter through an interactive process of the request information and the above indication information.

Optionally, the third parameter is a preconfigured parameter.

In a possible implementation of the first aspect, the first communications device performs noise perturbation processing on model parameters of the first AI model based on the first information to obtain updated model parameters of the first AI model, including: when a second condition is met, the first communications device performs noise perturbation processing on the model parameters of the first AI model based on the first information to obtain updated model parameters of the first AI model;

The second condition includes at least one of the following:

The training accuracy of the at least one AI model is lower than an eighth threshold;

The test accuracy of the at least one AI model is lower than a ninth threshold;

The system performance of the AI model system in which the at least one AI model resides is lower than a tenth threshold;

The system performance of the communication system in which the communication device deploys the at least one AI model is located is lower than an eleventh threshold;

The change in the neural network input and output distribution of the at least one AI model is greater than a twelfth threshold;

The neural network weight distribution of the at least one AI model is greater than a thirteenth threshold;

The change in the neural network weight of the at least one AI model is lower than a fourteenth threshold.

Based on the above solution, when the second condition is met, the first communication device can determine that the current operation of the first AI model may be abnormal. To this end, the first communication device can process the model parameters of the first AI model to obtain updated model parameters of the first AI model to update the first AI model. In this way, the problem of the first AI model's plasticity being reduced or even eliminated during long-term continuous learning can be alleviated or avoided, and the adaptability and robustness of continuous learning can be improved.

In a possible implementation of the first aspect, the first communication device obtains the first information, including: the first communication device receives the first information; or, the first communication device obtains measurement parameters of the first AI model and/or the second AI model, and determines the first information based on the measurement parameters of the first AI model and/or the second AI model.

Based on the above solution, the first communication device can obtain the first information through the above multiple methods to improve the flexibility of implementing the solution.

The second aspect of the present application provides a communication method, which is applied to a first communication device, which may be a communication device (such as a terminal device or a network device), or the first communication device may be a partial component in the communication device (such as a processor, a chip or a chip system, etc.), or the first communication device may also be a logic module or software that can realize all or part of the functions of the communication device. In this method, the first communication device obtains first information, which is used to indicate performance monitoring information of at least one AI model, and the at least one AI model is associated with a second AI model for deployment in the second communication device; the first communication device sends second information based on the first information, which is obtained by performing noise perturbation processing based on the first parameter, and the second information is used to update the second AI model.

Based on the above solution, after the first communication device obtains the performance monitoring information of at least one AI model through the first information, the first communication device can send the second information obtained by adding noise and perturbation based on the first parameter, so that the recipient of the second information (such as the second communication device) can obtain the second information obtained by adding noise and perturbation based on the first parameter, and update the second AI model based on the second information. In this way, the problem of the second AI model's plasticity being reduced or even disappearing during long-term continuous learning can be slowed down or avoided, and the adaptability of continuous learning can be improved, and the robustness of continuous learning can be enhanced.

In a possible implementation of the second aspect, the at least one AI model includes the first AI model and/or the second AI model; wherein the first AI model is associated with the second AI model, and the second AI model is used to be deployed on the second communication device.

Based on the above solution, the first information is used to indicate performance monitoring information of at least one AI model, and when the performance monitoring information indicated by the first information is used to trigger a certain condition (such as the first condition and/or the second condition described below), the first information can be used to perform noise perturbation processing on the model parameters of the first AI model and/or the second AI model. Accordingly, the at least one AI model may include the first AI model and/or the second AI model. In this way, the first communication device can trigger noise perturbation processing of the AI model based on the performance of the associated AI model.

Optionally, the first AI model is associated with the second AI model, including any of the following:

The first AI model is a public model and the second AI model is a personalized model;

The first AI model is a personalized model and the second AI model is a public model;

The input of the first AI model includes the output of the second AI model;

The input of the second AI model includes the output of the first AI model;

The first AI model is a teacher model and the second AI model is a student model; or,

The first AI model is a student model and the second AI model is a teacher model.

In a possible implementation manner of the second aspect, the second information is obtained by processing a quantization result of the first parameter, or the second information is obtained by processing a transmission parameter of the first parameter.

Based on the above solution, the second information can be obtained by performing noise disturbance processing in a variety of ways to improve the flexibility of the solution implementation.

Optionally, the first parameter includes one or more of the following:

The updated model parameters of the second AI model, used to update the gradient information of the second AI model, used to update the intermediate feature information of the second AI model, used to update the intermediate gradient information of the second AI model, the inference result of the first AI model, the distillation loss information corresponding to the second AI model, and the task loss information of the second AI model.

In a possible implementation of the second aspect, the first communication device sending the second information based on the first information includes: when a first condition is met, the first communication device sending the second information based on the first information;

The first condition includes at least one of the following:

The training accuracy of the at least one AI model is lower than a first threshold;

The test accuracy of the at least one AI model is lower than a second threshold;

The system performance of the AI model system in which the at least one AI model resides is lower than a third threshold;

The system performance of the communication system where the communication device deploying the at least one AI model is located is lower than a fourth threshold;

A change in the neural network input-output distribution of the at least one AI model is greater than a fifth threshold;

The neural network weight distribution of the at least one AI model is greater than a sixth threshold;

The change in the neural network weight of the at least one AI model is lower than a seventh threshold.

Based on the above solution, when the first condition is met, the first communication device can determine that the current operation of the second AI model may be abnormal. To this end, the first communication device can send second information, so that the second communication device can obtain the second information obtained by noise perturbation processing based on the first parameter, and update the second AI model based on the second information. In this way, the problem of the second AI model's plasticity being reduced or even disappearing during long-term continuous learning can be alleviated or avoided, and the adaptability and robustness of continuous learning can be improved.

In a possible implementation manner of the second aspect, the method further includes: the first communication device receiving indication information indicating a second parameter, where the second parameter is used to process the first parameter to obtain the second information.

Based on the above scheme, the first communication device can also determine the second parameter through the received indication information, and process the first parameter based on the second parameter to obtain the second information, that is, the first communication device can determine the second parameter based on the indication of other communication devices (for example, the recipient of the second information), so that the recipient of the second information can obtain the corresponding second information through the specified second parameter.

It should be understood that the second parameter can be used to perform noise disturbance processing on the first parameter to obtain the second information. Accordingly, the second parameter can be understood as a noise disturbance parameter, a noise parameter, or a disturbance parameter, etc.

In a possible implementation manner of the second aspect, the method further includes: the first communication device sending request information for requesting the second parameter.

Based on the above solution, the first communication device can determine the second parameter through an interactive process of the request information and the above indication information.

Optionally, the second parameter is a preconfigured parameter.

In a possible implementation of the second aspect, the first communication device obtains the first information, including: the first communication device receives the first information; or, the first communication device obtains measurement parameters of the first AI model and/or the second AI model, and determines the first information based on the measurement parameters of the first AI model and/or the second AI model.

Based on the above solution, the first communication device can obtain the first information through the above multiple methods to improve the flexibility of implementing the solution.

The third aspect of the present application provides a communication method, which is applied to a second communication device, which can be a communication device (such as a terminal device or a network device), or the second communication device can be a partial component in the communication device (such as a processor, a chip or a chip system, etc.), or the second communication device can also be a logic module or software that can implement all or part of the functions of the communication device. In this method, the second communication device receives second information, which is obtained by performing noise perturbation processing based on the first parameter; wherein the second information is used to update a second AI model deployed in the second communication device; the second communication device updates the second AI model based on the second information.

Based on the above solution, the second information received by the second communication device is obtained by performing noise perturbation processing based on the first parameter, and the second communication device can update the second AI model based on the second information. In other words, in the process of updating the second AI model by the second communication device, the second information used to update the second AI model is obtained by performing noise perturbation processing based on the first parameter. In this way, the problem of reduced or even disappearance of the plasticity of the second AI model during long-term continuous learning can be alleviated or avoided, and the adaptability of continuous learning can be improved, and the robustness of continuous learning can be enhanced.

In a possible implementation manner of the third aspect, the second information is obtained by processing a quantization result of the first parameter, or the second information is obtained by processing a transmission parameter of the first parameter.

Based on the above solution, the second information can be obtained by performing noise disturbance processing in a variety of ways to improve the flexibility of the solution implementation.

Optionally, the first parameter includes one or more of the following:

The updated model parameters of the second AI model, used to update the gradient information of the second AI model, used to update the intermediate feature information of the second AI model, used to update the intermediate gradient information of the second AI model, the inference result of the first AI model, the distillation loss information corresponding to the second AI model, and the task loss information of the second AI model.

In a fourth aspect, the present application provides a communication device, which is a first communication device and includes a transceiver unit; the processing unit is used to obtain first information, where the first information is used to indicate performance monitoring information of at least one AI model, and the at least one AI model is associated with a first AI model deployed in the first communication device; the processing unit is also used to perform noise disturbance processing on the model parameters of the first AI model based on the first information to obtain updated model parameters of the first AI model.

In the fourth aspect of the present application, the constituent modules of the communication device can also be used to execute the steps performed in each possible implementation method of the first aspect and achieve corresponding technical effects. For details, please refer to the first aspect and will not be repeated here.

In a fifth aspect, the present application provides a communication device, which is a first communication device and includes a transceiver unit and a processing unit; the processing unit is used to obtain first information, and the first information is used to indicate performance monitoring information of at least one AI model, and the at least one AI model is associated with a second AI model for deployment in a second communication device; the transceiver unit is used to send second information based on the first information, and the second information is obtained by noise disturbance processing based on the first parameter, and the second information is used to update the second AI model.

In the fifth aspect of this application, the constituent modules of the communication device can also be used to execute the steps performed in each possible implementation method of the second aspect and achieve corresponding technical effects. For details, please refer to the second aspect and will not be repeated here.

In a sixth aspect, the present application provides a communication device, which is a second communication device and includes a transceiver unit and a processing unit. The transceiver unit is used to receive second information, and the second information is obtained by performing noise disturbance processing based on the first parameter; wherein the second information is used to update a second AI model deployed in the second communication device; and the processing unit is used to update the second AI model based on the second information.

In the sixth aspect of this application, the constituent modules of the communication device can also be used to execute the steps performed in each possible implementation method of the third aspect and achieve corresponding technical effects. For details, please refer to the third aspect and will not be repeated here.

In a seventh aspect, the present application provides a communication device, comprising at least one processor coupled to a memory; the memory is configured to store programs or instructions; the at least one processor is configured to execute the programs or instructions, so that the communication device implements the method described in any possible implementation of any one of the first to third aspects. Optionally, the communication device may include the memory.

In an eighth aspect, the present application provides a communication device comprising at least one logic circuit and an input/output interface; the logic circuit is used to execute the method described in any possible implementation of any one of the first to third aspects.

In a ninth aspect, the present application provides a communication system, which includes the above-mentioned first communication device and second communication device.

In a tenth aspect, the present application provides a computer-readable storage medium for storing one or more computer-executable instructions. When the computer-executable instructions are executed by a processor, the processor executes the method described in any possible implementation of any one of the first to third aspects above.

In an eleventh aspect of the present application, a computer program product (or computer program) is provided. When the computer program in the computer program product is executed by the processor, the processor executes the method described in any possible implementation of any one of the first to third aspects above.

A twelfth aspect of the present application provides a chip system, which includes at least one processor for supporting a communication device to implement the method described in any possible implementation of any one of the first to third aspects above.

In one possible design, the 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 a chip or may include a chip and other discrete components. Optionally, the chip system may further include an interface circuit for providing program instructions and/or data to the at least one processor.

Among them, the technical effects brought about by any design method in the fourth to twelfth aspects can refer to the technical effects brought about by the different design methods in the above-mentioned first to third aspects, and will not be repeated here.

BRIEF DESCRIPTION OF THE DRAWINGS

Figures 1a to 1c are schematic diagrams of a communication system provided by this application;

Figures 2a to 2g are schematic diagrams of the AI processing process involved in this application;

FIG3 is an interactive schematic diagram of the communication method provided by this application;

Figures 4a to 4c are schematic diagrams of the model processing provided by this application;

FIG5 is an interactive diagram of the communication method provided by this application;

6 to 10 are schematic diagrams of the communication device provided in this application.

DETAILED DESCRIPTION

First, some of the terms used in the embodiments of the present application are explained to facilitate understanding by those skilled in the art.

(1) Terminal device: It 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.

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 (also known as "cellular" phones, mobile phones), computers, and data cards. For example, they can be portable, pocket-sized, handheld, computer-built-in, or vehicle-mounted mobile devices that exchange voice and/or data with the radio access network. For example, personal communication service (PCS) phones, cordless phones, session initiation protocol (SIP) phones, wireless local loop (WLL) stations, personal digital assistants (PDAs), tablets (tablets or pads), computers with wireless transceiver capabilities, and other devices. Wireless terminal equipment can also be called system, subscriber unit, subscriber station, mobile station/mobile station (MS), remote station, access point (AP), remote terminal (REMOTE terminal), access terminal (ACCESS terminal), user terminal (USER terminal), user agent (USER agent), subscriber station (SS), customer premises equipment (CPE), terminal, user equipment (UE), mobile terminal (MT), etc.

As an example and not a limitation, in the embodiments of the present application, the terminal device may also be a wearable device. Wearable devices may also be referred to as wearable smart devices or smart wearable devices, etc., which are a general term for wearable devices that are intelligently designed and developed using wearable technology for daily wear, such as glasses, gloves, watches, clothing, and shoes. A wearable device is a portable device that is worn directly on the body or integrated into the user's clothes or accessories. Wearable devices are not only hardware devices, but also achieve powerful functions through software support, data interaction, and cloud interaction. Broadly speaking, wearable smart devices include those that are fully functional, large in size, and can achieve complete or partial functions without relying on smartphones, such as smart watches or smart glasses, etc., as well as those that only focus on a certain type of application function and need to be used in conjunction with other devices such as smartphones, such as various smart bracelets, smart helmets, and smart jewelry for vital sign monitoring.

The terminal may also be a drone, a robot, a terminal in device-to-device (D2D) communication, a terminal in vehicle to everything (V2X), a virtual reality (VR) terminal device, an augmented reality (AR) terminal device, a wireless terminal in industrial control, a wireless terminal in self-driving, a wireless terminal in telemedicine (telehealth services), a wireless terminal in a smart grid, a wireless terminal in transportation safety, a wireless terminal in a smart city, a wireless terminal in a smart home, etc.

Furthermore, the terminal device may also be a terminal device in a communication system evolved after the fifth-generation (5G) communication system (e.g., 5G Advanced or sixth-generation (6G) communication system), or a terminal device in a future-evolved public land mobile network (PLMN). For example, 5G Advanced or 6G networks may further expand the form and functionality of 5G communication terminals. 6G terminals include, but are not limited to, vehicles, cellular network terminals (with integrated satellite terminal functionality), drones, and Internet of Things (IoT) devices.

In an embodiment of the present application, the terminal device may also obtain artificial intelligence (AI) services provided by the network device. Optionally, the terminal device may also have AI processing capabilities.

(2) Network equipment: It can be a device in a wireless network. For example, a network device can be a RAN node (or device) that connects a terminal device to a wireless network, which can also be called a base station. Currently, some examples of RAN equipment are: base station, evolved NodeB (eNodeB), gNB (gNodeB) in a 5G communication system, transmission reception point (TRP), evolved NodeB (eNB), radio network controller (RNC), NodeB (NB), home base station (e.g., home evolved NodeB, or home NodeB, HNB), baseband unit (BBU), or wireless fidelity (Wi-Fi) access point (AP). In addition, in a network structure, a network device can include a central unit (CU) node, a distributed unit (DU) node, or a RAN device including a CU node and a DU node.

Alternatively, a RAN node can be a macro base station, micro base station, indoor base station, relay node, donor node, or wireless controller in a cloud radio access network (CRAN) scenario. A RAN node can also be a server, wearable device, vehicle, or onboard device. For example, the access network device in vehicle-to-everything (V2X) technology can be a roadside unit (RSU).

In another possible scenario, multiple RAN nodes collaborate to assist terminals in achieving wireless access, with different RAN nodes implementing portions of the base station's functionality. For example, a RAN node can be a CU, DU, CU-control plane (CP), CU-user plane (UP), or radio unit (RU). The CU and DU can be separate or included in the same network element, such as a baseband unit (BBU). The RU can be included in a radio frequency device or radio unit, such as a remote radio unit (RRU), an active antenna unit (AAU), a radio head (RH), or a remote radio head (RRH).

In different systems, CU (or CU-CP and CU-UP), DU or RU may have different names, but those skilled in the art can understand their meanings. For example, in an open access network (open RAN, O-RAN or ORAN) system, CU may also be called O-CU (open CU), DU may also be called O-DU, CU-CP may also be called O-CU-CP, CU-UP may also be called O-CU-UP, and RU may also be called O-RU. For the convenience of description, this application uses CU, CU-CP, CU-UP, DU and RU as examples for description. Any unit among the CU (or CU-CP, CU-UP), DU and RU in this application can be implemented by a software module, a hardware module, or a combination of a software module and a hardware module.

Communication between access network equipment 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. 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.

For the correspondence between network elements in the ORAN system and their achievable protocol layer functions, please refer to Table 1 below.

Table 1

The network device may be any other device that provides wireless communication functionality to the terminal device. The embodiments of this application do not limit the specific technology and device form used by the network device. For ease of description, the embodiments of this application do not limit this.

The 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 the fourth generation (4G) network; and the access and mobility management function (AMF), user plane function (UPF), or session management function (SMF) in the 5G network. In addition, the core network equipment may also include other core network equipment in the 5G network and the next generation network of the 5G network.

In an embodiment of the present application, the above-mentioned network device may also have a network node with AI capabilities, which can provide AI services for terminals or other network devices. For example, it can be an AI node on the network side (access network or core network), a computing power node, a RAN node with AI capabilities, a core network element with AI capabilities, etc.

In the embodiments of the present application, the apparatus for implementing the function of the network device may be a network device, or may be a device capable of supporting the network device in implementing the function, such as a chip system, which may be provided in the network device. In the technical solutions provided in the embodiments of the present application, the technical solutions provided in the embodiments of the present application are described by taking the network device as an example of the apparatus for implementing the function of the network device.

(3) Configuration and pre-configuration: In this application, configuration and pre-configuration are used simultaneously. Configuration refers to the network device/server sending some parameter configuration information or parameter values to the terminal through messages or signaling, so that the terminal can determine the communication parameters or resources during transmission based on these values or information. Pre-configuration is similar to configuration, and can be parameter information or parameter values pre-negotiated between the network device/server and the terminal device, or parameter information or parameter values used by the base station/network device or terminal device as specified in the standard protocol, or parameter information or parameter values pre-stored in the base station/server or terminal device. This application does not limit this.

Furthermore, these values and parameters can be changed or updated.

(4) The terms "system" and "network" in the embodiments of the present application can be used interchangeably. "Multiple" refers to two or more. "And/or" describes the association relationship of associated objects, indicating that three relationships can exist. For example, A and/or B can mean: A exists alone, A and B exist at the same time, and B exists alone, where A and B can be singular or plural. The character "/" generally indicates that the previous and next associated objects are in an "or" relationship. "At least one of the following" or similar expressions refers 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. In addition, unless otherwise specified, the ordinal numbers such as "first" and "second" mentioned in the embodiments of the present application are used to distinguish multiple objects, and are not used to limit the order, timing, priority or importance of multiple objects.

(5) “Sending” and “receiving” in the embodiments of the present application indicate the direction of signal transmission. For example, “sending information to XX” can be understood as the destination of the information being XX, which can include direct sending through the air interface, as well as indirect sending through the air interface by other units or modules. “Receiving information from YY” can be understood as the source of the information being YY, which can include direct receiving from YY through the air interface, as well as indirect receiving from YY through the air interface from other units or modules. “Sending” can also be understood as the “output” of the chip interface, and “receiving” can also be understood as the “input” of the chip interface.

In other words, sending and receiving can be performed between devices, for example, between a network device and a terminal device, or can be performed within a device, for example, sending or receiving between components, modules, chips, software modules or hardware modules within the device through a bus, wiring or interface.

It is understandable that information may be processed between the source and destination of information transmission, such as coding, modulation, etc., but the destination can understand the valid information from the source. Similar expressions in this application can be understood similarly and will not be repeated.

(6) In the embodiments of the present application, "indication" may include direct indication and indirect indication, and may also include explicit indication and implicit indication. The information indicated by a certain information (such as the indication information described below) is called information to be indicated. In the specific implementation process, there are many ways to indicate the information to be indicated, such as but not limited to, directly indicating the information to be indicated, such as the information to be indicated itself or the index of the information to be indicated. The information to be indicated may also be indirectly indicated by indicating other information, wherein the other information is associated with the information to be indicated; or only a part of the information to be indicated may be indicated, while the other part of the information to be indicated is known or agreed in advance. For example, the indication of specific information may be achieved by means of the arrangement order of each information agreed in advance (such as predefined by the protocol), thereby reducing the indication overhead to a certain extent. The present application does not limit the specific method of indication. It is understandable that for the sender of the indication information, the indication information can be used to indicate the information to be indicated, and for the receiver of the indication information, the indication information can be used to determine the information to be indicated.

In this application, unless otherwise specified, the same or similar parts between the various embodiments can refer to each other. In the various embodiments of this application, and the various methods/designs/implementations in each embodiment, if there is no special explanation and logical conflict, the terms and/or descriptions between different embodiments and the various methods/designs/implementations in each embodiment are consistent and can be referenced to each other. The technical features in different embodiments and the various methods/designs/implementations in each embodiment can be combined to form new embodiments, methods, or implementations according to their inherent logical relationships. The following description of the implementation methods of this application does not constitute a limitation on the scope of protection of this application.

This application can be applied to a long-term evolution (LTE) system, a new radio (NR) system, or a communication system evolved after 5G (e.g., 6G). The communication system includes at least one network device and/or at least one terminal device.

Please refer to Figure 1a, which is a schematic diagram of a communication system in this application. Figure 1a exemplarily illustrates a network device and six terminal devices, namely terminal device 1, terminal device 2, terminal device 3, terminal device 4, terminal device 5, and terminal device 6. In the example shown in Figure 1a, terminal device 1 is a smart teacup, terminal device 2 is a smart air conditioner, terminal device 3 is a smart gas pump, terminal device 4 is a vehicle, terminal device 5 is a mobile phone, and terminal device 6 is a printer.

As shown in Figure 1a, the sending entity of AI configuration information can be a network device. The receiving entity of AI configuration information can be terminal devices 1-6. In this case, the network device and terminal devices 1-6 form a communication system. In this communication system, terminal devices 1-6 can send data to the network device, and the network device needs to receive data sent by terminal devices 1-6. At the same time, the network device can send configuration information to terminal devices 1-6.

For example, in Figure 1a, terminal devices 4 and 6 can also form a communication system. Terminal device 5 acts as a network device, i.e., the sending entity of AI configuration information; terminal devices 4 and 6 act as terminal devices, i.e., the receiving entities of AI configuration information. For example, in a connected vehicle system, terminal device 5 sends AI configuration information to terminal devices 4 and 6, respectively, and receives data from terminal devices 4 and 6. Correspondingly, terminal devices 4 and 6 receive AI configuration information from terminal device 5 and send data to terminal device 5.

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

As shown in Figure 1b, taking the 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.

As shown in Figure 1c, taking the terminal devices including a TV and a mobile phone as an example, communication-related services and AI-related services can also be performed between the TV and the mobile phone.

The technical solution provided in this application can be applied to wireless communication systems (for example, the systems shown in Figures 1a, 1b, or 1c). For example, an AI network element can be introduced into the communication system provided in this application to implement some or all AI-related operations. The AI network element can also be referred to as an AI node, AI device, AI entity, AI module, AI model, or AI unit, etc. The AI network element can be a network element built into the communication system. For example, the AI network element can be an AI module built into: an access network device, a core network device, a cloud server, or an operation, administration, and maintenance (OAM) system to implement AI-related functions. The OAM can serve as a network manager for a core network device and/or as a network manager for an access network device. Alternatively, the AI network element can also be an independently set network element in the communication system. Optionally, the terminal or the chip built into the terminal can also include an AI entity to implement AI-related functions.

The following is a brief introduction to the AI that may be involved in this application.

AI can imbue machines with human intelligence. For example, it can use computer hardware and software to simulate certain intelligent human behaviors. Machine learning methods can be used to achieve artificial intelligence. In machine learning, a machine uses training data to learn (or train) a model. This model represents the mapping from input to output. The learned model can be used for inference (or prediction), meaning that the model can be used to predict the output corresponding to a given input. This output can also be called an inference result (or prediction result).

Machine learning can include supervised learning, unsupervised learning, and reinforcement learning. Among them, unsupervised learning can also be called unsupervised learning.

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

Unsupervised learning uses algorithms to discover inherent patterns in collected sample values. One type of unsupervised learning algorithm uses the samples themselves as supervisory signals, meaning the model learns the mapping from one sample to another. This is called self-supervised learning. During training, the model parameters are optimized by calculating the error between the model's predictions and the samples themselves. Self-supervised learning can be used in signal compression and decompression recovery applications. Common algorithms include autoencoders and generative adversarial networks.

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 lack explicit label data for "correct" actions. Instead, the algorithm must interact with the environment to obtain reward signals from the environment, and then adjust its decision-making actions to maximize the reward signal value. For example, in downlink power control, the reinforcement learning model adjusts the downlink transmit power of each user based on the overall system throughput fed back by the wireless network, hoping to achieve higher system throughput. The goal of reinforcement learning is also to learn the mapping between environmental states and optimal (e.g., optimal) decision-making actions. However, because the labels for "correct actions" cannot be obtained in advance, network optimization cannot be achieved by calculating the error between actions and "correct actions." Reinforcement learning training is achieved through iterative interaction with the environment.

A neural network (NN) is a specific model in machine learning technology. According to the universal approximation theorem, NNs can theoretically approximate any continuous function, enabling them to learn arbitrary mappings. Traditional communication systems require extensive expert knowledge to design communication modules. However, deep learning communication systems based on neural networks can automatically discover implicit patterns in massive data sets and establish mapping relationships between data, achieving performance superior to traditional modeling methods.

The idea of a neural network is derived from the neuronal structure of the brain. For example, each neuron performs a weighted sum operation on its input values and outputs the result through an activation function.

As shown in Figure 2a, it is a schematic diagram of a neuron structure. Assume that the input of the neuron is x = [x ₀ , x ₁ ,…, x _n ], and the weights corresponding to each input are w = [w ₀ , w ₁ ,…, w _n ], where n is a positive integer, and w _i and _xi can be various possible types such as decimals, integers (such as 0, positive integers or negative integers, etc.), or complex numbers. w _i is used as the weight of _xi to weight _xi . The bias for weighted summation of input values according to the weights is, for example, b. There can be many forms of activation functions. Assuming that the activation function of a neuron is: y = f(z) = max(0,z), the output of the neuron is: For another example, if the activation function of a neuron is: y = f(z) = z, then the output of the neuron is: b can be a decimal, an integer (eg, 0, a positive integer, or a negative integer), or a complex number, etc. The activation functions of different neurons in a neural network can be the same or different.

Furthermore, neural networks generally include 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 comprises, and the number of neurons in each layer can be referred to as the width of that layer. In one implementation, a neural network includes an input layer and an output layer. The input layer processes the input information received by the neural network through neurons, passing the processing results to the output layer, which then obtains the output of the neural network. In another implementation, a neural network includes an input layer, a hidden layer, and an output layer. The input layer processes the input information received by the neural network through neurons, passing the processing results to an intermediate hidden layer. The hidden layer performs calculations on the received processing results to obtain a calculation result, which is then passed to the output layer or the next adjacent hidden layer, which ultimately obtains the output of the neural network. A neural network can include one hidden layer or multiple hidden layers connected in sequence, without limitation.

An example of a neural network is a deep neural network (DNN). Depending on how the network is constructed, DNNs can include feedforward neural networks (FNNs), convolutional neural networks (CNNs), and recurrent neural networks (RNNs).

Figure 2b is a schematic diagram of an FNN network. A characteristic of FNN networks is that neurons in adjacent layers are fully connected. This characteristic typically requires a large amount of storage space and results in high computational complexity.

CNN is a type of neural network specifically designed to process data with a grid-like structure. For example, time series data (e.g., discrete sampling on the time axis) and image data (e.g., discrete sampling on the two-dimensional axis) can both be considered grid-like data. CNNs do not utilize all input information at once for computation. Instead, they use a fixed-size window to intercept a portion of the information for convolution operations, significantly reducing the computational complexity of model parameters. Furthermore, depending on the type of information intercepted by the window (e.g., people and objects in an image represent different types of information), each window can use a different convolution kernel, enabling CNNs to better extract features from the input data.

RNNs are a type of DNN that utilizes feedback time series information. RNN inputs include the current input value and its own output value at the previous moment. RNNs are suitable for capturing temporally correlated sequence features and are particularly well-suited for applications such as speech recognition and channel coding.

During the machine learning model training process, a loss function can be defined. This function describes the gap or discrepancy between the model's output and the desired target value. Loss functions can be expressed in a variety of forms, and their specific form is not restricted. The model training process can be viewed as adjusting some or all of the model's parameters to keep the loss function below a threshold or meet the target.

A model may also be referred to as an AI model, rule, or other name. An AI model can be considered a specific method for implementing an AI function. An AI model represents a mapping relationship or function between the input and output of a model. AI functions may include one or more of the following: data collection, model training (or model learning), model information release, model inference (or model reasoning, inference, or prediction, etc.), model monitoring or model verification, or inference result release, etc. AI functions may also be referred to as AI (related) operations, or AI-related functions.

The following is an illustrative description of the implementation process of the fully connected neural network with reference to the accompanying drawings.

1. Fully connected neural network, also called multilayer perceptron (MLP).

As shown in Figure 2c, an MLP consists of an input layer (left), an output layer (right), and multiple hidden layers (center). Each layer of the MLP contains several nodes, called neurons. Neurons in adjacent layers are connected to each other.

Alternatively, considering neurons in two adjacent layers, the output h of the neurons in the next layer is the weighted sum of all neurons x in the previous layer connected to it and passes through the activation function, which can be expressed as:
h＝f(wx+b).

Among them, w is the weight matrix, b is the bias vector, and f is the activation function.

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

Where m is the index of the neural network layer, m is greater than or equal to 1, and m is less than or equal to M, where M is the total number of neural network layers.

In other words, a neural network can be understood as a mapping from an input data set to an output data set. Neural networks are typically initialized randomly, and the process of obtaining this mapping from random w and b using existing data is called neural network training.

Optionally, a specific method of training is to use a loss function to evaluate the output results of the neural network.

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 loss function reaches a minimum, which is the "better point (e.g., optimal point)" in Figure 2d. It is understood that the neural network parameters corresponding to the "better point (e.g., optimal point)" in Figure 2d can be used as the neural network parameters in the trained AI model information.

Alternatively, the gradient descent process can be expressed as:

Among them, θ is the parameter to be optimized (including w and b), L is the loss function, and η is the learning rate, which controls the step size of gradient descent. represents the derivative operation, represents the derivative of θ with respect to L.

Optionally, the backpropagation process utilizes the chain rule for partial derivatives.

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

Among them, _wij is the weight of node j connecting to node i, and _si is the weighted sum of the inputs on node i.

2. Federated learning (FL).

The concept of federated learning effectively solves the current difficulties faced by the development of artificial intelligence. On the premise of fully protecting user data privacy and security, it efficiently completes the model learning task by promoting the collaboration between various edge devices and central servers.

As shown in Figure 2f, the FL architecture is the most widely used training architecture in the current FL field. The FedAvg algorithm is the basic algorithm of FL. Its algorithm flow is roughly as follows:

(1) The center initializes the model to be trained And broadcast it to all client devices.

(2) In the round t∈[1,T], client k∈[1,K] based on the local dataset For the received global model Perform E epochs of training to obtain local training results Report it to the central node.

(3) The central node aggregates and collects the local training results from all (or some) clients. Assume that the client set that uploads the local model in round t is The center will use the number of samples of the corresponding client as the weight to perform weighted averaging to obtain a new global model. The specific update rule is: The center then sends the latest version of the global model Broadcast to all client devices for a new round of training.

(4) Repeat steps (2) and (3) until the model finally converges or the number of training rounds reaches the upper limit.

In addition to reporting local models You can also use the local gradient of training After reporting, the central node averages the local gradients and updates the global model according to the direction of the average gradient.

As you can see, in the FL framework, datasets exist on distributed nodes. Distributed nodes collect local datasets, perform local training, and report the local training results (models or gradients) to the central node. The central node itself does not have a dataset; it is only responsible for fusing the training results of distributed nodes to obtain a global model and send it to the distributed nodes.

3. Decentralized learning: Different from federated learning, decentralized learning is another distributed learning architecture.

As shown in Figure 2g, consider a fully distributed system without a central node. The design goal f(x) of a decentralized learning system is generally the mean of the goals _fi (x) of each node, that is, Where p is the number of distributed nodes, x is the parameter to be optimized. In machine learning, x is the parameter of the machine learning (such as neural network) model. Each node uses local data and local target _fi (x) to calculate the local gradient Then it is sent to the neighboring nodes that can be communicated with. After any node receives the gradient information sent by its neighbor, it can update the parameter x of the local model according to the following formula:

in, represents the parameters of the local model after the k+1th (k is a natural number) update in the i-th node, Represents the parameters of the local model after the kth update in the i-th node (if k is 0, it means (where αk is the parameter of the local model of node i that is not updated), _αk represents the tuning coefficient, _Pi is the set of neighboring nodes of node i, and | _Pi | represents the number of elements in the set of neighboring nodes of node i, that is, the number of neighboring nodes of node i. Through information exchange between nodes, the decentralized learning system will eventually learn a unified model.

The technical solutions provided in this application can be applied to wireless communication systems (e.g., the systems shown in Figures 1a, 1b, or 1c). In wireless communication systems, communication nodes generally have both signal transceiver capabilities and computing capabilities. For example, network devices with computing capabilities primarily provide computing power to support signal transceiver capabilities (e.g., performing signal transmission and reception processing) to enable communication between the network device and other communication nodes.

However, in communication networks, communication nodes may have excess computing power beyond just supporting the aforementioned communication tasks. Therefore, how to utilize this computing power is a pressing technical issue.

In one possible implementation, a communication node can serve as a participating node in an AI learning system, and the computing power of the communication node can be applied to a certain link of the AI learning system (e.g., the AI learning system described in FIG2f or FIG2g). Generally speaking, in an AI learning system, an AI node can establish a data set according to a task, complete the training of the model offline, and perform reasoning after online deployment. However, after a long period of continuous learning, the AI model may have the problem of disappearing plasticity, that is, the AI model will not be able to learn the knowledge of new training samples, resulting in the performance of the AI model becoming worse and worse, resulting in a decrease in the robustness of continuous learning. As a possible optimization method, a method of selectively reinitializing some parameters during training can be used to ensure that the neural network can maintain the ability to learn new samples. However, in this optimization method, it is necessary to design complex functions and count the life cycles of various parameters to select the reset parameter set, which is highly complex; at the same time, reinitializing the parameters will greatly affect the instantaneous performance of the neural network, resulting in the above-mentioned optimization method not being able to effectively solve the above-mentioned problem.

In order to solve the above problems, the present application provides a communication method and related equipment, which will be described in detail below with reference to the accompanying drawings.

Please refer to FIG3 , which is a schematic diagram of an implementation of the communication method provided in this application. The method includes the following steps.

It should be noted that, in the following, FIG3 and FIG5 illustrate the method by taking the first communication device and other communication devices (e.g., the second communication device) as the execution subjects of the interaction diagram as examples, but the present application does not limit the execution subjects of the interaction diagram. For example, the communication device can be a communication device (e.g., a terminal device or a network device), or a chip, a baseband chip, a modem chip, a system-on-chip (SoC) chip including a modem core, a system-in-package (SIP) chip, a communication module, a chip system, a processor, a logic module, or software in the communication device.

S301. A second communication device sends first information, and a first communication device receives the first information accordingly. The first information is used to indicate performance monitoring information of at least one AI model associated with a first AI model deployed on the first communication device.

S302. The first communication device performs noise disturbance processing on the model parameters of the first AI model based on the first information to obtain updated model parameters of the first AI model.

In this application, AI model can be replaced by other terms, such as neural network, neural network model, AI neural network model, machine learning model, or AI processing model.

In this application, noise disturbance processing can be replaced by other terms, such as noise addition processing, disturbance processing, noise disturbance processing, noise interference processing, or random disturbance processing.

In one possible implementation, the first information acquired by the first communication device is used to indicate performance monitoring information of at least one AI model, wherein the at least one AI model includes the first AI model and/or the second AI model, the first AI model is associated with the second AI model, and the second AI model is used to be deployed on the second communication device. The first information can be used to indicate the performance monitoring information of the at least one AI model, and in S302, the first communication device can perform noise perturbation processing on the model parameters of the first AI model based on the first information to obtain updated model parameters of the first AI model. In other words, when the performance monitoring information indicated by the first information meets a certain condition (e.g., the first condition and/or the second condition described below), the first information can be used to trigger noise perturbation processing of the model parameters of the first AI model and/or the second AI model. Accordingly, the at least one AI model can include the first AI model and/or the second AI model. In this way, the first communication device can trigger noise perturbation processing of the AI model based on the performance of the associated AI model.

It should be noted that S301 is optional. For the first communication device, in addition to obtaining the first information through interaction with the second communication device, the first communication device can also obtain the first information through other means. Exemplarily, the first communication device can obtain the first information locally. For example, the first communication device can obtain measurement parameters of the first AI model and/or the second AI model, and determine the first information based on the measurement parameters of the first AI model and/or the second AI model.

As an example, the measured parameters of an AI model (e.g., the first AI model and/or the second AI model) may include the output of the AI model. In other words, the first communications device may determine performance monitoring information of the AI model based on the output of the AI model. For example, the performance monitoring information may be determined by a mathematical relationship (e.g., at least one of a difference, variance, standard deviation, etc.) between the output of the AI model and a preconfigured tag.

As another example, when the output of an AI model (e.g., the first AI model and/or the second AI model) is a communication parameter (e.g., at least one of the encoding and/or decoding code rate, the sequence used for scrambling and/or descrambling, the modulation and/or demodulation order, the channel parameter, the communication rate, the bit error rate, etc.), the measurement parameter of the AI model may include communication performance information for communication based on the communication parameter (e.g., one or more of the bit error rate, the block error rate, the reference signal received power (RSRP), and the signal to interference plus noise ratio (SINR)). In other words, the first communication device may determine the performance monitoring information of the AI model based on the communication performance information and the preconfigured performance information. For example, the performance monitoring information may be determined by a mathematical relationship (e.g., at least one of the difference, variance, standard deviation, etc.) between the communication performance information and the preconfigured performance information.

Accordingly, the first communication device can obtain the measurement parameters of the AI model in a variety of ways. For example, the first communication device can obtain the measurement parameters of the first AI model based on the output of the local first AI model. In another example, the first communication device can obtain the measurement parameters of the first AI model based on the communication process with other communication devices. In another example, the first communication device can obtain the measurement parameters of the second AI model based on the communication process with a second communication device.

In a possible implementation, the first AI model is associated with the second AI model, including any one of the following methods A to F:

Method A: The first AI model is a public model and the second AI model is a personalized model.

Method B: The first AI model is a personalized model and the second AI model is a public model.

Method C: The input of the first AI model includes the output of the second AI model.

Mode D: The input of the second AI model includes the output of the first AI model.

Method E: The first AI model is a teacher model and the second AI model is a student model.

Method F: The first AI model is a student model and the second AI model is a teacher model.

Methods A and B can be understood as the federated learning scenario shown in Figure 2f. In federated learning, there are central nodes and distributed nodes. The central node stores the public model and periodically aggregates the personalized models uploaded by the distributed nodes. The distributed nodes periodically download the public model and fine-tune it using local data to obtain personalized models. In Method A, the first communication device can be the central node and the second communication device can be a distributed node. In Method B, the second communication device can be the central node and the first communication device can be a distributed node.

As an example, as shown in Figure 4a, it is a schematic diagram of an implementation of federated learning, taking the network device as the central node and the terminal device as the distributed node as an example. Among them, the network device can save the public model, and the terminal device 1 to the terminal device 3 can save their own personalized models. The two can interact through the model download process and the model upload process. Among them, some or all AI models in the terminal device can participate in federated learning. In the example shown in Figure 4a, the AI models of terminal device 1 and terminal device 2 can all participate in federated learning, and the AI model of terminal device 3 can include a part filled with a black pattern that participates in federated learning, and a part filled with a blank that does not participate in federated learning.

Each device shown in Figure 4a can serve as a first communication device to perform noise perturbation on the model parameters of a locally deployed AI model. In other words, by adding noise perturbations to the model parameters of a public or personalized model, the robustness of continuous federated learning to distributional changes is increased.

As an example of noise perturbation processing, for any AI model, the standard deviation of the noise perturbation σ _i can be defined, and the model parameter set ( represents a set, _pi represents an element in the set), _pi = 1 represents adding noise perturbation to the parameter (i ranges from 0 to N-1, N is the number of parameters of the neural network parameters involved in the noise perturbation), and the value is (i.e., the noise n _i obeys Normal distribution, mean 0, variance ), p _i = 0 means no disturbance is required.

Optionally, the parameters for the noise perturbation process may be part or all of the model parameters of the AI model, for example, the part or all of the model parameters are determined by the above The set is determined. For example, if the AI model has 1000 (i.e. N=1000) parameters, The set is {1,0,0,1…}, The sequence of 1000 elements in the collection corresponds to these 1000 parameters, The model parameters corresponding to the position where the value of _pi in the set is 1 are subjected to noise disturbance processing.

As another example of noise perturbation processing, for any model parameter set The AI model of the collection, noise disturbance processing satisfies:
e _i =e _i +n _i ;

Where, e _i represents the i-th parameter of the model. When p _i = 1, (i.e., the noise n _i obeys Normal distribution, mean 0, variance ); when p _i =0, n _i =0.

Alternatively, in addition to the _normal distribution, other implementations are possible. _For example, a and b represent the lower and upper bounds.

Methods C and D can be understood as split learning scenarios. In split learning, the AI model can be split into two or more parts.

As an example, Figure 4b shows a schematic diagram of split learning. In this example, the AI model is split into two parts, which can be an encoder and a decoder. The encoder and decoder communicate intermediate features and intermediate gradients. During training, data enters the encoding neural network, obtains intermediate features, and transmits them to the decoding neural network. The decoding neural network calculates the loss based on the label values and updates the decoding neural network in reverse order. Simultaneously, the decoding neural network transmits the intermediate gradients to the encoding neural network, which then updates the encoding neural network based on the received intermediate gradients.

The encoder or decoder shown in Figure 4b can be used as the first AI model in the first communication device to perform noise perturbation processing. In other words, noise can be added to some or all of the model parameters of the encoder or decoder, thereby increasing the robustness of continuous split learning to distribution changes. Similarly, for the noise perturbation method for some or all of the model parameters of the encoder or decoder, please refer to the description of the previous example.

Methods E and F can be understood as knowledge distillation scenarios. Knowledge distillation can enhance the loss of smaller or simpler models (such as student models or knowledge learning models) through the output of larger or more complex models (such as teacher models or knowledge transfer models), thereby achieving better convergence performance or faster convergence speed. This can be considered a model compression technology.

As an example, Figure 4c is a schematic diagram of knowledge distillation. This example takes the models involved in the knowledge distillation process as an example, including a teacher model and a student model. Generally speaking, the teacher model does not participate in training, but this is not limited here, that is, there is no distinction between specific teacher models and student models. The teacher model can also be guided by the student model to update, or the two models can be updated at the same time, or there can be multiple teacher models or multiple student models. The following only takes the most common single teacher guiding single student model learning as an example. In Figure 4c, the teacher model and the student model respectively infer and obtain inference results. According to the inference results of the teacher model and the student model, the knowledge distillation loss (KD loss) can be obtained. The device deployed with the student model can obtain the task loss (task Loss) according to the label, and then the device can update the model parameters of the student model according to the distillation loss and task loss.

The teacher model or student model shown in FIG4c can be used as the first AI model in the first communication device to perform noise perturbation processing. In other words, noise can be added to some or all model parameters of the teacher model or student model, thereby increasing the robustness of the continuous knowledge distillation process to distribution changes. Similarly, the noise perturbation method for some or all model parameters of the teacher model or student model can refer to the description of the previous example.

In one possible implementation, in S302, when the second condition is met, the first communication device performs noise perturbation processing on the model parameters of the first AI model based on the first information to obtain updated model parameters of the first AI model; wherein the second condition includes at least one of the following:

The training accuracy of the at least one AI model is lower than a first threshold;

The test accuracy of the at least one AI model is lower than a second threshold;

The system performance of the AI model system in which the at least one AI model resides is lower than a third threshold;

The system performance of the communication system where the communication device deploying the at least one AI model is located is lower than a fourth threshold;

A change in the neural network input-output distribution of the at least one AI model is greater than a fifth threshold;

The neural network weight distribution of the at least one AI model is greater than a sixth threshold;

The change in the neural network weight of the at least one AI model is lower than a seventh threshold.

Optionally, the first communication device may determine the values of the first to seventh thresholds in a variety of ways. For example, one or more of the first to seventh thresholds may be preconfigured. In another example, one or more of the first to seventh thresholds may be configured for the first communication device by another device (e.g., a second communication device, a network device, or a server).

When the second condition is met, the first communication device may determine that the current operation of the first AI model may be abnormal. To this end, the first communication device may process the model parameters of the first AI model to obtain updated model parameters of the first AI model to update the first AI model. In this way, the problem of the first AI model's plasticity being reduced or even disappearing during long-term continuous learning can be alleviated or avoided, and the adaptability and robustness of continuous learning can be improved.

Based on the solution shown in Figure 3, after the first communication device obtains performance monitoring information of at least one AI model through the first information, in S302, the first communication device can perform noise perturbation processing on the model parameters of the first AI model to obtain updated model parameters of the first AI model. In this way, the computing power of the communication node can be applied to AI model processing, and the AI model can be updated based on the performance monitoring information of the AI model.

In addition, in S302, the process of the first communication device performing the AI model parameter update is triggered based on the performance monitoring information of at least one AI model, and during the AI model parameter update process, the first communication device obtains the updated AI model parameters through noise perturbation processing. In this way, the solution can be applied to the scenario of continuous learning, and by performing noise perturbation processing on the AI model parameters, the problem of reduced or even disappearance of the plasticity of the AI model during long-term continuous learning can be slowed down or avoided, which can improve the adaptability of continuous learning and enhance the robustness of continuous learning.

In a possible implementation of the method shown in FIG3 , before S302, the method further includes: the first communication device receives indication information indicating a third parameter, and the third parameter is used to process the model parameters of the first AI model to obtain the updated model parameters of the first AI model. Specifically, the first communication device can also receive the third parameter determined by the received indication information, and perform noise perturbation processing on the model parameters of the first AI model based on the third parameter, that is, the first communication device can determine the third parameter based on the indication of other communication devices (such as the recipient of the second information), so that the first communication device can obtain the noise perturbation processing through the specified third parameter.

Exemplarily, the second communication device may send indication information indicating the third parameter to the first communication device based on the first information, that is, the second communication device may send indication information indicating the third parameter under the condition that it is determined that the above-mentioned second condition is met, so that the first communication device obtains the third parameter and executes S302.

Optionally, the indication information indicating the third parameter may include at least one of the third parameter and an index of the third parameter (the index may be implemented in a manner shown in Table 2 or Table 3 below).

It should be understood that the third parameter can be used to process the model parameters of the first AI model to obtain the updated model parameters of the first AI model. Accordingly, the third parameter can be understood as a noise disturbance parameter, a noise parameter, or a disturbance parameter. For example, the third parameter can be the set described above. σ, One or more of .

Optionally, before the first communication device receives the indication information indicating the third parameter, the method further includes: the first communication device sending request information for requesting the third parameter. Thus, the first communication device can determine the third parameter through an interaction between the request information and the indication information. Exemplarily, the first communication device can send the request information for requesting the third parameter based on the first information. That is, the first communication device can send the request information for requesting the third parameter upon determining that the second condition is met, thereby obtaining the third parameter and executing S302.

Optionally, the third parameter is a preconfigured parameter.

As an example, the third parameter includes the set For example, the set This can be determined from Table 2 below.

Table 2

As an example, taking the third parameter including σ as an example, σ can be determined by the following Table 3.

Table 3

As can be seen from Figure 3 and the related description above, the first communication device can perform noise perturbation processing on the model parameters of the first AI model deployed on the first communication device based on the performance monitoring information of at least one AI model to improve the robustness of continuous learning. The at least one AI model can be associated with the first AI model or associated with the second AI model deployed on the second communication device. In other words, the first communication device can also perform noise perturbation processing on the model parameters of the second AI model deployed on the second communication device based on the performance monitoring information of at least one AI model to improve the robustness of continuous learning. The process shown in Figure 5 will be described below.

Please refer to FIG5 , which is another schematic diagram of the communication method provided in this application.

S501. A second communication device sends first information, and a first communication device receives the first information accordingly. The first information is used to indicate performance monitoring information of at least one AI model associated with a first AI model deployed on the first communication device.

It should be noted that S501 is optional. For the first communication device, in addition to obtaining the first information through interaction with the second communication device, the first communication device can also obtain the first information through other means. For the specific implementation process, please refer to the above S301 and related descriptions.

S502. The first communication device sends second information based on the first information, and the second communication device receives the second information accordingly. The second information is obtained by performing noise perturbation processing based on the first parameter, and the second information is used to update a second AI model; the second AI model is deployed on the second communication device.

Optionally, the first parameter includes one or more of the following: model parameters of the updated second AI model, gradient information used to update the second AI model, intermediate feature information used to update the second AI model, intermediate gradient information used to update the second AI model, inference results of the first AI model, distillation loss information corresponding to the second AI model, and task loss information of the second AI model.

In a possible implementation, the second information sent by the first communication device in S502 may be implemented in a variety of ways, which will be described in detail below.

In a first approach, the second information is obtained by processing a quantized result of the first parameter.

Optionally, in method one, when the second information is obtained by processing based on the quantization result of the first parameter, after the second communication device receives the second information in S502, the second communication device can perform analysis processing on the received second information (for example, the analysis processing may include one or more of channel equalization, demodulation processing, decoding processing, digital domain processing, etc.), and update the second AI model based on the result of the analysis processing.

In a second approach, the second information is obtained by processing the transmission parameters of the first parameter.

Optionally, in the second method, when the second information is obtained by processing the transmission parameter of the first parameter, the implementation process can be understood as an over-the-air calculation process, that is, after the second communication device receives the second information in S502, the second communication device may not perform digital domain processing on the received second information, but directly update the second AI model based on the original information of the received second information. For example, the first communication device converts one or more parameter values contained in the first parameter into one or more first symbols; after performing noise perturbation processing on the one or more first symbols to obtain one or more second symbols, the one or more second symbols are mapped to air interface resources (e.g., at least one of time domain resources, frequency domain resources, and air domain resources) to generate the second information. In this example, the first communication device does not perform source encoding on the first parameter at the application layer and then send it back to the physical layer as a bit stream, which is then channel-coded and symbol-modulated by the physical layer and then generated and sent as a signal, but converts the first parameter into one or more symbols before transmitting. Therefore, the method of transmitting the first parameter on the wireless air interface can increase the transmission speed of the first parameter, thereby improving communication efficiency, and increasing the processing speed of the second communication device for model update based on the second information.

Optionally, the transmission parameters of the first parameter may include one or more of a transmit power parameter, a precoding parameter, and an indication of whether to use method 2 for processing (the indication information can be used to control whether one or more communication devices use method 2 to transmit information, so as to control the signal-to-noise ratio of the signal after superposition in the air by controlling the number of accessed communication devices).

The following will take method 2 as an example to exemplify the process of sending the second information under different implementation methods of the AI model.

As an example, when the first AI model and the second AI model are implemented by the method shown in method A or method B above, the first parameter may include the model parameter of the updated second AI model, which is used to update one or more items of the gradient information of the second AI model.

For example, taking the process shown in FIG4a as an example, the first communication device may be a network device. The first communication device may, during the process of issuing the public model (i.e., the first parameters include the model parameters or update gradients of the public model), perform noise perturbation processing on the model parameters or update gradients of the public model to obtain second information, and transmit the second information in S502.

For another example, using the process shown in FIG4a as an example, the first communication device can be any terminal device. During the process of uploading the personalized model (i.e., the first parameters include model parameters or update gradients of the personalized model), the first communication device can perform noise perturbation processing on the model parameters or update gradients of the personalized model to obtain second information, and transmit the second information in S502.

In other words, during the model download or upload process, lossless transmission can be changed to lossy transmission, that is, the model parameters or update gradients will be superimposed with transmission noise during uploading and downloading to achieve the effect of noise disturbance processing.

As another example, when the first AI model and the second AI model are implemented by the method shown in Method C or Method D above, the first parameter may include intermediate feature information for updating the second AI model, and one or more items of intermediate gradient information for updating the second AI model.

Similarly, taking the process shown in Figure 4b as an example, the first communication device can add noise disturbance when transmitting the intermediate features and intermediate gradients, so that the transmission mode of the intermediate features and intermediate gradients of the encoder or decoder is changed from lossless transmission to lossy transmission, that is, the intermediate features and intermediate gradients will be superimposed with transmission noise during uploading/downloading to achieve the effect of noise disturbance processing.

As another example, when the first AI model and the second AI model are implemented by the method shown in Method E or Method F above, the first parameter may include one or more of the inference result of the first AI model, the distillation loss information corresponding to the second AI model, and the task loss information of the second AI model.

Similarly, using the process shown in Figure 4c as an example, assume that the formula for calculating the distillation loss is loss _KD = L( _ft , _fs ), where _ft and _fs are the inputs to the teacher model and student model, respectively, when calculating the distillation loss. These inputs can be intermediate features of the AI model or the output of the AI model. During knowledge distillation training, the distillation loss can be calculated on the same node as the teacher model, on the same node as the student model, or separately on a different node. Therefore, for knowledge distillation, lossy transmission can be used to transmit _ft and/or _fs to achieve the effect of noise perturbation.

In a possible implementation, in S502, when a first condition is met, the first communication device sends the second information based on the first information; wherein the first condition includes at least one of the following:

The training accuracy of the at least one AI model is lower than an eighth threshold;

The test accuracy of the at least one AI model is lower than a ninth threshold;

The system performance of the AI model system in which the at least one AI model resides is lower than a tenth threshold;

The system performance of the communication system in which the communication device deploys the at least one AI model is located is lower than an eleventh threshold;

The change in the neural network input and output distribution of the at least one AI model is greater than a twelfth threshold;

The neural network weight distribution of the at least one AI model is greater than a thirteenth threshold;

The change in the neural network weight of the at least one AI model is lower than a fourteenth threshold.

Optionally, the first condition and the second condition may be the same, for example, the i-th (i is 1 to 7) threshold value from the first to the seventh threshold value is the same as the i-th threshold value from the eighth to the fourteenth threshold value.

Optionally, the first condition and the second condition may be different, for example, the i-th (i is 1 to 7) threshold value from the first to seventh threshold values is partially different or completely different from the i-th threshold value from the eighth to fourteenth threshold values.

Optionally, the first communication device may determine the values of the eighth to fourteenth thresholds in a variety of ways. For example, one or more of the eighth to fourteenth thresholds may be preconfigured. In another example, one or more of the eighth to fourteenth thresholds may be configured for the first communication device by another device (e.g., a second communication device, a network device, or a server).

Among them, when the first condition is met, the first communication device can determine that the current operation of the second AI model may be abnormal. To this end, the first communication device can send second information, so that the second communication device can obtain second information obtained by noise perturbation processing based on the first parameter, and update the second AI model based on the second information. In this way, the problem of the second AI model's plasticity being reduced or even disappearing during long-term continuous learning can be alleviated or avoided, and the adaptability of continuous learning can be improved, and the robustness of continuous learning can be enhanced.

Based on the scheme shown in Figure 5, in S502, the first communication device can send the second information obtained by performing noise perturbation processing based on the first parameter based on the first information, so that the recipient of the second information (such as the second communication device) can obtain the second information obtained by performing noise perturbation processing based on the first parameter, and update the second AI model based on the second information. In this way, compared with the method of directly sending the first parameter without performing noise perturbation processing, the problem of reduced or even disappearance of the plasticity of the second AI model during long-term continuous learning can be slowed down or avoided, and the adaptability of continuous learning can be improved, and the robustness of continuous learning can be enhanced.

It should be noted that the method shown in FIG5 can be combined with the method shown in FIG3 above. For example, after S301 shown in FIG3, S502 shown in FIG5 can also be executed. The execution order of S302 and S502 is not limited. That is, S302 can be executed first and then S502, or S502 can be executed first and then S302.

In one possible implementation of the method shown in FIG5 , before S502, the method further includes: the first communication device receiving indication information indicating a second parameter, the second parameter being used to process the first parameter to obtain the second information. Specifically, the first communication device may also determine the second parameter based on the received indication information, and process the first parameter based on the second parameter to obtain the second information. That is, the first communication device may determine the second parameter based on an indication from another communication device (e.g., a recipient of the second information), so that the recipient of the second information can obtain the corresponding second information using the specified second parameter.

Exemplarily, the second communication device may send indication information indicating the second parameter to the first communication device based on the first information, that is, the second communication device may send indication information indicating the second parameter under the condition that it is determined that the above-mentioned first condition is met, so that the first communication device obtains the second parameter and executes S502.

Optionally, the indication information indicating the second parameter may include at least one of the second parameter and an index of the second parameter (the index may be implemented in a manner as shown in Table 2 or Table 3 above).

It should be understood that the second parameter can be used to perform noise disturbance processing on the first parameter to obtain the second information. Accordingly, the second parameter can be understood as a noise disturbance parameter, a noise parameter, or a disturbance parameter, etc.

Optionally, before the first communication device receives the indication information indicating the second parameter, the method further includes: the first communication device sending request information for requesting the second parameter. Thus, the first communication device can determine the second parameter through an interaction between the request information and the indication information. Exemplarily, the first communication device can send the request information for requesting the second parameter based on the first information. That is, the first communication device can send the request information for requesting the second parameter upon determining that the first condition is met, thereby obtaining the second parameter and executing S502.

Optionally, the second parameter is a preconfigured parameter.

Referring to FIG. 6 , an embodiment of the present application provides a communication device 600. This communication device 600 can implement the functions of the first communication device (or second communication device) in the above-described method embodiment, thereby also achieving the beneficial effects of the above-described method embodiment. In this embodiment of the present application, the communication device 600 can be the first communication device (or second communication device), or it can be an integrated circuit or component within the first communication device (or second communication device), such as a chip, a baseband chip, a modem chip, a SoC chip including a modem core, a system-in-package (SIP) chip, a communication module, a chip system, a processor, etc.

It should be noted that the transceiver unit 602 may include a sending unit and a receiving unit, which are respectively used to perform sending and receiving.

In one possible implementation, when the communication device 600 is used to execute the method executed by the first communication device in Figure 3 and related embodiments, the communication device 600 includes a processing unit 601; the processing unit 601 is used to obtain first information, which is used to indicate performance monitoring information of at least one AI model, and the at least one AI model is associated with the first AI model deployed in the first communication device; the processing unit 601 is also used to perform noise disturbance processing on the model parameters of the first AI model based on the first information to obtain updated model parameters of the first AI model.

In one possible implementation, when the communication device 600 is used to execute the method executed by the first communication device in Figure 5 and related embodiments, the communication device 600 includes a processing unit 601; the processing unit 601 is used to obtain first information, and the first information is used to indicate performance monitoring information of at least one AI model, and the at least one AI model is associated with a second AI model for deployment in a second communication device; the transceiver unit 602 is used to send second information based on the first information, and the second information is obtained by noise disturbance processing based on the first parameter, and the second information is used to update the second AI model.

In one possible implementation, when the communication device 600 is used to execute the method executed by the second communication device in Figure 5 and related embodiments, the communication device 600 includes a processing unit 601; the transceiver unit 602 is used to receive second information, which is obtained by noise disturbance processing based on the first parameter; wherein the second information is used to update the second AI model deployed in the second communication device; the processing unit 601 is used to update the second AI model based on the second information.

In one possible design, when the communication device 600 is a terminal device or a communication module in a terminal, the functions of the processing unit 601 can be implemented by one or more processors. Specifically, the processor can include a modem chip, or a SoC chip or SIP chip containing a modem core. The functions of the transceiver unit 602 can be implemented by a transceiver circuit.

In one possible design, when the communication device 600 is a circuit or chip responsible for communication functions in a terminal, such as a modem chip or a SoC chip or SIP chip containing a modem core, the functions of the processing unit 601 can be implemented by a circuit system including one or more processors or processor cores in the above chip. The functions of the transceiver unit 602 can be implemented by an interface circuit or data transceiver circuit on the above chip.

It should be noted that, for details on the information execution process of the units of the above-mentioned communication device 600, please refer to the description in the method embodiment shown above in this application, and no further details will be given here.

Please refer to Fig. 7, which is another schematic structural diagram of a communication device 700 provided in this application. The communication device 700 includes a logic circuit 701 and an input/output interface 702. The communication device 700 may be a chip or an integrated circuit.

The transceiver unit 602 shown in Figure 6 may include a communication interface, which may include the input/output interface 702 in Figure 7 , which may include an input interface and an output interface. Alternatively, the communication interface may be a transceiver circuit, which may include an input interface circuit and an output interface circuit.

In one possible implementation, when the communication device 700 is used to execute the method executed by the first communication device in Figure 3 and related embodiments, the logic circuit 701 is used to obtain first information, where the first information is used to indicate performance monitoring information of at least one AI model, and the at least one AI model is associated with the first AI model deployed in the first communication device; the logic circuit 701 is also used to perform noise disturbance processing on the model parameters of the first AI model based on the first information to obtain updated model parameters of the first AI model.

In one possible implementation, when the communication device 700 is used to execute the method executed by the first communication device in Figure 5 and related embodiments, the logic circuit 701 is used to obtain first information, which is used to indicate performance monitoring information of at least one AI model, and the at least one AI model is associated with a second AI model deployed in a second communication device; the input and output interface 702 is used to send second information based on the first information, which is obtained by noise disturbance processing based on the first parameter, and the second information is used to update the second AI model.

In one possible implementation, when the communication device 700 is used to execute the method executed by the second communication device in Figure 5 and related embodiments, the input-output interface 702 is used to receive second information, which is obtained by noise disturbance processing based on the first parameter; wherein the second information is used to update the second AI model deployed in the second communication device; and the logic circuit 701 is used to update the second AI model based on the second information.

The logic circuit 701 and the input/output interface 702 may also execute other steps executed by the first communication device or the second communication device in any embodiment and achieve corresponding beneficial effects, which will not be described in detail here.

In a possible implementation, the processing unit 601 shown in FIG. 6 may be the logic circuit 701 in FIG. 7 .

Optionally, the logic circuit 701 may be a processing device, and the functions of the processing device may be partially or entirely implemented by software. The functions of the processing device may be partially or entirely implemented by software.

Optionally, the processing device 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 corresponding processing and/or steps in any one of the method embodiments.

Alternatively, the processing device may include only a processor. A memory for storing the computer program is located outside the processing device, and the processor is connected to the memory via circuits/wires to read and execute the computer program stored in the memory. The memory and processor may be integrated or physically separate.

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 processing circuits (DSPs), microcontrollers (MCUs), programmable logic devices (PLDs), or other integrated chips, or any combination of the above chips or processors.

Please refer to Figure 8, which shows the communication device 800 involved in the above-mentioned embodiments provided in an embodiment of the present application. The communication device 800 can specifically be a communication device serving as a terminal device in the above-mentioned embodiments. The example shown in Figure 8 is that the terminal device is implemented through the terminal device (or a component in the terminal device).

Herein, a possible logical structure diagram of the communication device 800 is shown. The communication device 800 may include but is not limited to at least one processor 801 and a communication port 802 .

The transceiver unit 602 shown in Figure 6 may include a communication interface, which may include the communication port 802 shown in Figure 8 , which may include an input interface and an output interface. Alternatively, the communication port 802 may also be a transceiver circuit, which may include an input interface circuit and an output interface circuit.

Further optionally, the communication device 800 may also include at least one of a memory 803 and a bus 804. In an embodiment of the present application, the at least one processor 801 is used to control and process the actions of the communication device 800.

Furthermore, the processor 801 may be a central processing unit (CPU), a general-purpose processor (GPPC), a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic device (PLD), a transistor logic device (TLD), a hardware component, or any combination thereof. It may implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. The processor may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, and the like. Those skilled in the art will clearly understand that, for ease and brevity of description, the specific operating processes of the systems, devices, and units described above may refer to the corresponding processes in the aforementioned method embodiments and will not be further described herein.

It should be noted that the communication device 800 shown in Figure 8 can be specifically used to implement the steps implemented by the terminal device in the aforementioned method embodiment and achieve the corresponding technical effects of the terminal device. The specific implementation methods of the communication device shown in Figure 8 can refer to the description in the aforementioned method embodiment and will not be repeated here.

Please refer to Figure 9, which is a structural diagram of the communication device 900 involved in the above-mentioned embodiments provided in an embodiment of the present application. The communication device 900 can specifically be a communication device as a network device in the above-mentioned embodiments. The example shown in Figure 9 is that the network device is implemented through the network device (or a component in the network device), wherein the structure of the communication device can refer to the structure shown in Figure 9.

The communication device 900 includes at least one processor 911 and at least one network interface 914. Further optionally, the communication device also includes at least one memory 912, at least one transceiver 913 and one or more antennas 915. The processor 911, the memory 912, the transceiver 913 and the network interface 914 are connected, for example, via a bus. In an embodiment of the present application, the connection may include various interfaces, transmission lines or buses, etc., which are not limited in this embodiment. The antenna 915 is connected to the transceiver 913. The network interface 914 is used to enable the communication device to communicate with other communication devices through a communication link. For example, the network interface 914 may include a network interface between the communication device and the core network device, such as an S1 interface, and the network interface may include a network interface between the communication device and other communication devices (such as other network devices or core network devices), such as an X2 or Xn interface.

The transceiver unit 602 shown in Figure 6 may include a communication interface, which may include the network interface 914 shown in Figure 9. The network interface 914 may include an input interface and an output interface. Alternatively, the network interface 914 may be a transceiver circuit, which may include an input interface circuit and an output interface circuit.

Processor 911 is primarily used to process communication protocols and communication data, control the entire communication device, execute software programs, and process software program data, for example, to support the communication device in performing the actions described in the embodiments. The communication device may include a baseband processor and a central processing unit. The baseband processor is primarily used to process communication protocols and communication data, while the central processing unit is primarily used to control the entire terminal device, execute software programs, and process software program data. Processor 911 in Figure 9 may integrate the functions of both a baseband processor and a central processing unit. Those skilled in the art will appreciate that the baseband processor and the central processing unit may also be independent processors interconnected via a bus or other technology. Those skilled in the art will appreciate that a terminal device may include multiple baseband processors to accommodate different network standards, multiple central processing units to enhance its processing capabilities, and various components of the terminal device may be connected via various buses. The baseband processor may also be referred to as a baseband processing circuit or a baseband processing chip. The central processing unit may also be referred to as a central processing circuit or a central processing chip. The functionality for processing communication protocols and communication data may be built into the processor or stored in memory as a software program, which is executed by the processor to implement the baseband processing functionality.

The memory is primarily used to store software programs and data. Memory 912 can exist independently and be connected to processor 911. Alternatively, memory 912 and processor 911 can be integrated together, for example, within a single chip. Memory 912 can store program code for executing the technical solutions of the embodiments of the present application, and execution is controlled by processor 911. The various computer program codes executed can also be considered drivers for processor 911.

Figure 9 shows only one memory and one processor. In an actual terminal device, there may be multiple processors and multiple memories. The memory may also be referred to as a storage medium or a storage device. The memory may be a storage element on the same chip as the processor, i.e., an on-chip storage element, or an independent storage element, which is not limited in the present embodiment.

The transceiver 913 can be used to support the reception or transmission of radio frequency signals between the communication device and the terminal, and the transceiver 913 can be connected to the antenna 915. The transceiver 913 includes a transmitter Tx and a receiver Rx. Specifically, one or more antennas 915 can receive radio frequency signals. The receiver Rx of the transceiver 913 is used to receive the radio frequency signal from the antenna, convert the radio frequency signal into a digital baseband signal or a digital intermediate frequency signal, and provide the digital baseband signal or digital intermediate frequency signal to the processor 911 so that the processor 911 can further process the digital baseband signal or digital intermediate frequency signal, such as demodulation and decoding. In addition, the transmitter Tx in the transceiver 913 is also used to receive a modulated digital baseband signal or digital intermediate frequency signal from the processor 911, convert the modulated digital baseband signal or digital intermediate frequency signal into a radio frequency signal, and send the radio frequency signal through one or more antennas 915. Specifically, the receiver Rx can selectively perform one or more stages of down-mixing and analog-to-digital conversion on the RF signal to obtain a digital baseband signal or a digital intermediate frequency signal. The order of the down-mixing and analog-to-digital conversion processes is adjustable. The transmitter Tx can selectively perform one or more stages of up-mixing and digital-to-analog conversion on the modulated digital baseband signal or digital intermediate frequency signal to obtain a RF signal. The order of the up-mixing and digital-to-analog conversion processes is adjustable. The digital baseband signal and the digital intermediate frequency signal may be collectively referred to as digital signals.

The transceiver 913 may also be referred to as a transceiver unit, a transceiver, a transceiver device, etc. Optionally, a device in the transceiver unit that implements a receiving function may be referred to as a receiving unit, and a device in the transceiver unit that implements a transmitting function may be referred to as a transmitting unit. That is, the transceiver unit includes a receiving unit and a transmitting unit. The receiving unit may also be referred to as a receiver, an input port, a receiving circuit, etc., and the transmitting unit may be referred to as a transmitter, a transmitter, or a transmitting circuit, etc.

It should be noted that the communication device 900 shown in Figure 9 can be specifically used to implement the steps implemented by the network device in the aforementioned method embodiment and achieve the corresponding technical effects of the network device. The specific implementation methods of the communication device 900 shown in Figure 9 can refer to the description in the aforementioned method embodiment and will not be repeated here.

Please refer to FIG10 , which is a schematic structural diagram of the communication device involved in the above-mentioned embodiment provided in an embodiment of the present application.

It can be understood that the communication device 100 includes, for example, modules, units, elements, circuits, or interfaces, which are appropriately configured together to implement the technical solutions provided in this application. The communication device 100 can be the terminal device or network device described above, or a component (such as a chip) in these devices, used to implement the method described in the following method embodiment. The communication device 100 includes one or more processors 101. The processor 101 can be a general-purpose processor or a dedicated processor. For example, it can be a baseband processor or a central processing unit. The baseband processor can be used to process communication protocols and communication data, and the central processing unit can be used to control the communication device (such as a RAN node, terminal, or chip, etc.), execute software programs, and process data of software programs.

Optionally, in one design, the processor 101 may include a program 103 (sometimes also referred to as code or instructions), which may be executed on the processor 101 to cause the communication device 100 to perform the methods described in the following embodiments. In yet another possible design, the communication device 100 includes circuitry (not shown in FIG10 ).

Optionally, the communication device 100 may include one or more memories 102 on which a program 104 (sometimes also referred to as code or instructions) is stored. The program 104 can be run on the processor 101, so that the communication device 100 executes the method described in the above method embodiment.

Optionally, the processor 101 and/or the memory 102 may include AI modules 107 and 108, which are used to implement AI-related functions. The AI module may be implemented through software, hardware, or a combination of software and hardware. For example, the AI module may include a wireless intelligent control (RIC) module. For example, the AI module may be a near-real-time RIC or a non-real-time RIC.

Optionally, data may be stored in the processor 101 and/or the memory 102. The processor and the memory may be provided separately or integrated together.

Optionally, the communication device 100 may further include a transceiver 105 and/or an antenna 106. The processor 101 may also be sometimes referred to as a processing unit, and controls the communication device (e.g., a RAN node or terminal). The transceiver 105 may also be sometimes referred to as a transceiver unit, a transceiver, a transceiver circuit, or a transceiver, and is configured to implement the transceiver functions of the communication device through the antenna 106.

The processing unit 601 shown in FIG6 may be the processor 101. The transceiver unit 602 shown in FIG6 may be a communication interface, which may be the transceiver 105 shown in FIG10 . The transceiver 105 may include an input interface and an output interface. Alternatively, the transceiver 105 may be a transceiver circuit, which may include an input interface circuit and an output interface circuit.

An embodiment of the present application further provides a computer-readable storage medium, which is used to store one or more computer-executable instructions. When the computer-executable instructions are executed by a processor, the processor executes the method described in the possible implementation methods of the first communication device or the second communication device in the aforementioned embodiment.

An embodiment of the present application also provides a computer program product (or computer program). When the computer program product is executed by the processor, the processor executes the method that may be implemented by the above-mentioned first communication device or second communication device.

An embodiment of the present application also provides a chip system, which includes at least one processor for supporting a communication device to implement the functions involved in the possible implementation methods of the above-mentioned communication device. Optionally, the chip system also includes an interface circuit, which provides program instructions and/or data to the at least one processor. In one possible design, the chip system may also include a memory, which is used to store the necessary program instructions and data for the communication device. The chip system can be composed of chips, or it can include chips and other discrete devices, wherein the communication device can specifically be the first communication device or the second communication device in the aforementioned method embodiment.

An embodiment of the present application further provides a communication system, wherein the network system architecture includes the first communication device and/or the second communication device in any of the above embodiments.

In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms. Whether a function is performed 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 to be beyond the scope of this application.

The units described as separate components may or may not be physically separate, and 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 these units may be selected to achieve the purpose of this embodiment according to actual needs.

In addition, the functional units in the various embodiments of the present 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 above-mentioned integrated unit can be implemented in the form of hardware or in the form of a software functional unit. If the integrated unit is implemented in the form of 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 the present application is essentially or the contributing part or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in the various embodiments of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, and other media that can store program code.

Claims (72)

A communication method, characterized in that the method uses a first communication device, and the method includes: Obtaining first information, where the first information is used to indicate performance monitoring information of at least one artificial intelligence (AI) model, where the at least one AI model is associated with a first AI model deployed on the first communication device; Noise disturbance processing is performed on the model parameters of the first AI model based on the first information to obtain updated model parameters of the first AI model. The method according to claim 1, further comprising: Second information is sent based on the first information, where the second information is obtained by performing noise disturbance processing based on the first parameter, and the second information is used to update a second AI model; wherein the second AI model is used to be deployed in a second communication device. The method according to claim 2 is characterized in that the second information is obtained by processing the quantization result of the first parameter, or the second information is obtained by processing the transmission parameter of the first parameter. The method according to claim 2 or 3, wherein the first parameter includes one or more of the following: The updated model parameters of the second AI model, used to update the gradient information of the second AI model, used to update the intermediate feature information of the second AI model, used to update the intermediate gradient information of the second AI model, the inference result of the first AI model, the distillation loss information corresponding to the second AI model, and the task loss information of the second AI model. The method according to any one of claims 2 to 4, wherein sending the second information based on the first information comprises: When a first condition is met, sending second information based on the first information; The first condition includes at least one of the following: The training accuracy of the at least one AI model is lower than a first threshold; The test accuracy of the at least one AI model is lower than a second threshold; The system performance of the AI model system in which the at least one AI model is located is lower than a third threshold; The system performance of the communication system where the communication device deploying the at least one AI model is located is lower than a fourth threshold; The change in the neural network input and output distribution of the at least one AI model is greater than a fifth threshold; The neural network weight distribution of the at least one AI model is greater than a sixth threshold; The change in the neural network weight of the at least one AI model is lower than a seventh threshold. The method according to any one of claims 2 to 5, further comprising: Indication information indicating the second parameter is received, where the second parameter is used to process the first parameter to obtain the second information. The method according to claim 6, further comprising: Sending a request message for requesting the second parameter. The method according to any one of claims 1 to 7, further comprising: Indication information indicating a third parameter is received, where the third parameter is used to process the model parameters of the first AI model to obtain updated model parameters of the first AI model. The method according to claim 8, further comprising: Sending request information for requesting the third parameter. The method according to any one of claims 1 to 9, wherein the performing noise perturbation processing on the model parameters of the first AI model based on the first information to obtain updated model parameters of the first AI model comprises: When the second condition is met, performing noise perturbation processing on the model parameters of the first AI model based on the first information to obtain updated model parameters of the first AI model; The second condition includes at least one of the following: The training accuracy of the at least one AI model is lower than an eighth threshold; The test accuracy of the at least one AI model is lower than a ninth threshold; The system performance of the AI model system in which the at least one AI model is located is lower than a tenth threshold; The system performance of the communication system in which the communication device deploying the at least one AI model is located is lower than an eleventh threshold; The change in the neural network input and output distribution of the at least one AI model is greater than a twelfth threshold; The neural network weight distribution of the at least one AI model is greater than a thirteenth threshold; The change in the neural network weight of the at least one AI model is lower than the fourteenth threshold. The method according to any one of claims 1 to 10, wherein obtaining the first information comprises: receiving the first information; or, Obtain measurement parameters of the first AI model and/or the second AI model, and determine the first information based on the measurement parameters of the first AI model. The method according to any one of claims 1 to 11, characterized in that the at least one AI model includes the first AI model and/or the second AI model; wherein the first AI model is associated with the second AI model, and the second AI model is used to be deployed on the second communication device. A communication method, characterized in that it is applied to a first communication device, the method comprising: Obtaining first information, the first information being used to indicate performance monitoring information of at least one artificial intelligence (AI) model, the at least one AI model being associated with a second AI model deployed on a second communication device; Second information is sent based on the first information, where the second information is obtained by performing noise disturbance processing based on the first parameter, and the second information is used to update the second AI model. The method according to claim 13 is characterized in that the second information is obtained by processing the quantization result of the first parameter, or the second information is obtained by processing the transmission parameter of the first parameter. The method according to claim 13 or 14, wherein the first parameter includes one or more of the following: The updated model parameters of the second AI model, used to update the gradient information of the second AI model, used to update the intermediate feature information of the second AI model, used to update the intermediate gradient information of the second AI model, the inference result of the first AI model, the distillation loss information corresponding to the second AI model, and the task loss information of the second AI model. The method according to any one of claims 13 to 15, wherein sending the second information based on the first information comprises: sending the second information based on the first information when a first condition is met; The first condition includes at least one of the following: The training accuracy of the at least one AI model is lower than a first threshold; The test accuracy of the at least one AI model is lower than a second threshold; The system performance of the AI model system in which the at least one AI model is located is lower than a third threshold; The system performance of the communication system where the communication device deploying the at least one AI model is located is lower than a fourth threshold; The change in the neural network input and output distribution of the at least one AI model is greater than a fifth threshold; The neural network weight distribution of the at least one AI model is greater than a sixth threshold; The change in the neural network weight of the at least one AI model is lower than a seventh threshold. The method according to any one of claims 13 to 16, further comprising: Indication information indicating the second parameter is received, where the second parameter is used to process the first parameter to obtain the second information. The method according to claim 17, further comprising: Sending a request message for requesting the second parameter. The method according to any one of claims 13 to 18, wherein obtaining the first information comprises: receiving the first information; or, Obtain measurement parameters of the first AI model and/or the second AI model, and determine the first information based on the measurement parameters of the first AI model. The method according to any one of claims 13 to 19, characterized in that the at least one AI model includes the first AI model and/or the second AI model; wherein the first AI model is associated with the second AI model, and the second AI model is used to be deployed on the second communication device. A communication method, characterized in that it is applied to a second communication device, the method comprising: receiving second information, where the second information is obtained by performing noise perturbation processing based on the first parameter; wherein the second information is used to update a second artificial intelligence (AI) model deployed in the second communication device; The second AI model is updated based on the second information. The method according to claim 21 is characterized in that the second information is obtained by processing the quantization result of the first parameter, or the second information is obtained by processing the transmission parameter of the first parameter. The method according to claim 21 or 22, wherein the first parameter includes one or more of the following: The updated model parameters of the second AI model, used to update the gradient information of the second AI model, used to update the intermediate feature information of the second AI model, used to update the intermediate gradient information of the second AI model, the inference result of the first AI model, the distillation loss information corresponding to the second AI model, and the task loss information of the second AI model. A first communication device, characterized by comprising a processing unit; The processing unit is configured to obtain first information, where the first information is used to indicate performance monitoring information of at least one artificial intelligence (AI) model, where the at least one AI model is associated with a first AI model deployed on the first communication device; Noise disturbance processing is performed on the model parameters of the first AI model based on the first information to obtain updated model parameters of the first AI model. The device according to claim 24, further comprising a transceiver unit; The transceiver unit is used to send second information based on the first information, where the second information is obtained by performing noise disturbance processing based on the first parameter, and the second information is used to update a second AI model; wherein the second AI model is used to be deployed in a second communication device. The device according to claim 25 is characterized in that the second information is obtained by processing the quantization result of the first parameter, or the second information is obtained by processing the transmission parameter of the first parameter. The device according to claim 25 or 26, wherein the first parameter includes one or more of the following: The updated model parameters of the second AI model, used to update the gradient information of the second AI model, used to update the intermediate feature information of the second AI model, used to update the intermediate gradient information of the second AI model, the inference result of the first AI model, the distillation loss information corresponding to the second AI model, and the task loss information of the second AI model. The device according to any one of claims 25 to 27, characterized in that The transceiver unit is further configured to: When a first condition is met, sending second information based on the first information; The first condition includes at least one of the following: The training accuracy of the at least one AI model is lower than a first threshold; The test accuracy of the at least one AI model is lower than a second threshold; The system performance of the AI model system in which the at least one AI model is located is lower than a third threshold; The system performance of the communication system where the communication device deploying the at least one AI model is located is lower than a fourth threshold; The change in the neural network input and output distribution of the at least one AI model is greater than a fifth threshold; The neural network weight distribution of the at least one AI model is greater than a sixth threshold; The change in the neural network weight of the at least one AI model is lower than a seventh threshold. The device according to any one of claims 25 to 28, characterized in that The transceiver unit is further configured to receive indication information indicating the second parameter, where the second parameter is used to process the first parameter to obtain the second information. The device according to claim 29, characterized in that The transceiver unit is further configured to send request information for requesting the second parameter. The device according to any one of claims 24 to 30, characterized in that The transceiver unit is further configured to receive indication information indicating a third parameter, where the third parameter is used to process the model parameters of the first AI model to obtain updated model parameters of the first AI model. The device according to claim 31, characterized in that The transceiver unit is further configured to send request information for requesting the third parameter. The device according to any one of claims 24 to 32, characterized in that The processing unit is further configured to: When the second condition is met, performing noise perturbation processing on the model parameters of the first AI model based on the first information to obtain updated model parameters of the first AI model; The second condition includes at least one of the following: The training accuracy of the at least one AI model is lower than an eighth threshold; The test accuracy of the at least one AI model is lower than a ninth threshold; The system performance of the AI model system in which the at least one AI model is located is lower than a tenth threshold; The system performance of the communication system in which the communication device deploying the at least one AI model is located is lower than an eleventh threshold; The change in the neural network input and output distribution of the at least one AI model is greater than a twelfth threshold; The neural network weight distribution of the at least one AI model is greater than a thirteenth threshold; The change in the neural network weight of the at least one AI model is lower than the fourteenth threshold. The device according to any one of claims 24 to 33, characterized in that The processing unit is further configured to receive the first information; or Obtain measurement parameters of the first AI model and/or the second AI model, and determine the first information based on the measurement parameters of the first AI model. The device according to any one of claims 24 to 34, characterized in that the at least one AI model includes the first AI model and/or the second AI model; wherein the first AI model is associated with the second AI model, and the second AI model is used to be deployed on the second communication device. A first communication device, characterized by comprising a processing unit; The processing unit is configured to obtain first information indicating performance monitoring information of at least one artificial intelligence (AI) model, where the at least one AI model is associated with a second AI model deployed on a second communication device; Second information is sent based on the first information, where the second information is obtained by performing noise disturbance processing based on the first parameter, and the second information is used to update the second AI model. The device according to claim 36 is characterized in that the second information is obtained by processing the quantization result of the first parameter, or the second information is obtained by processing the transmission parameter of the first parameter. The device according to claim 36 or 37, wherein the first parameter includes one or more of the following: The updated model parameters of the second AI model, used to update the gradient information of the second AI model, used to update the intermediate feature information of the second AI model, used to update the intermediate gradient information of the second AI model, the inference result of the first AI model, the distillation loss information corresponding to the second AI model, and the task loss information of the second AI model. The device according to any one of claims 36 to 38, wherein the transceiver unit is further configured to: When a first condition is met, sending second information based on the first information; The first condition includes at least one of the following: The training accuracy of the at least one AI model is lower than a first threshold; The test accuracy of the at least one AI model is lower than a second threshold; The system performance of the AI model system in which the at least one AI model is located is lower than a third threshold; The system performance of the communication system where the communication device deploying the at least one AI model is located is lower than a fourth threshold; The change in the neural network input and output distribution of the at least one AI model is greater than a fifth threshold; The neural network weight distribution of the at least one AI model is greater than a sixth threshold; The change in the neural network weight of the at least one AI model is lower than a seventh threshold. The device according to any one of claims 36 to 39, characterized in that The transceiver unit is further configured to receive indication information indicating the second parameter, where the second parameter is used to process the first parameter to obtain the second information. The device according to claim 40, characterized in that The transceiver unit is further configured to send request information for requesting the second parameter. The device according to any one of claims 36 to 41, characterized in that The processing unit is further configured to receive the first information; or Obtain measurement parameters of the first AI model and/or the second AI model, and determine the first information based on the measurement parameters of the first AI model. The device according to any one of claims 36 to 42, characterized in that the at least one AI model includes the first AI model and/or the second AI model; wherein the first AI model is associated with the second AI model, and the second AI model is used to be deployed on the second communication device. A second communication device, characterized by comprising a processing unit and a transceiver unit; The transceiver unit is configured to receive second information, where the second information is obtained by performing noise perturbation processing based on the first parameter; wherein the second information is used to update a second artificial intelligence (AI) model deployed in the second communication device; The processing unit is configured to update the second AI model based on the second information. The device according to claim 44 is characterized in that the second information is obtained by processing the quantization result of the first parameter, or the second information is obtained by processing the transmission parameter of the first parameter. The device according to claim 44 or 45, wherein the first parameter includes one or more of the following: The updated model parameters of the second AI model, used to update the gradient information of the second AI model, used to update the intermediate feature information of the second AI model, used to update the intermediate gradient information of the second AI model, the inference result of the first AI model, the distillation loss information corresponding to the second AI model, and the task loss information of the second AI model. A first communication device, comprising a processor; The processor is configured to obtain first information, where the first information is used to indicate performance monitoring information of at least one artificial intelligence (AI) model, where the at least one AI model is associated with a first AI model deployed on the first communication device; Noise disturbance processing is performed on the model parameters of the first AI model based on the first information to obtain updated model parameters of the first AI model. The device according to claim 47, characterized in that the device further comprises a communication port; The communication port is further used to send second information based on the first information, where the second information is obtained by performing noise disturbance processing based on the first parameter, and the second information is used to update a second AI model; wherein the second AI model is used to be deployed in a second communication device. The device according to claim 48 is characterized in that the second information is obtained by processing the quantization result of the first parameter, or the second information is obtained by processing the transmission parameter of the first parameter. The device according to claim 48 or 49, wherein the first parameter includes one or more of the following: The updated model parameters of the second AI model, used to update the gradient information of the second AI model, used to update the intermediate feature information of the second AI model, used to update the intermediate gradient information of the second AI model, the inference result of the first AI model, the distillation loss information corresponding to the second AI model, and the task loss information of the second AI model. The device according to any one of claims 48 to 50, wherein the communication port is further used to: When a first condition is met, sending second information based on the first information; The first condition includes at least one of the following: The training accuracy of the at least one AI model is lower than a first threshold; The test accuracy of the at least one AI model is lower than a second threshold; The system performance of the AI model system in which the at least one AI model is located is lower than a third threshold; The system performance of the communication system where the communication device deploying the at least one AI model is located is lower than a fourth threshold; The change in the neural network input and output distribution of the at least one AI model is greater than a fifth threshold; The neural network weight distribution of the at least one AI model is greater than a sixth threshold; The change in the neural network weight of the at least one AI model is lower than a seventh threshold. The device according to any one of claims 48 to 51, characterized in that The communication port is further used to receive indication information indicating the second parameter, and the second parameter is used to process the first parameter to obtain the second information. The device according to claim 52, characterized in that The communication port is further used to send request information for requesting the second parameter. The device according to any one of claims 47 to 53, characterized in that The communication port is further used to receive indication information indicating a third parameter, where the third parameter is used to process the model parameters of the first AI model to obtain updated model parameters of the first AI model. The device according to claim 54, characterized in that The communication port is further used to send request information for requesting the third parameter. The device according to any one of claims 47 to 55, characterized in that The processor is further configured to: When the second condition is met, performing noise perturbation processing on the model parameters of the first AI model based on the first information to obtain updated model parameters of the first AI model; The second condition includes at least one of the following: The training accuracy of the at least one AI model is lower than an eighth threshold; The test accuracy of the at least one AI model is lower than a ninth threshold; The system performance of the AI model system in which the at least one AI model is located is lower than a tenth threshold; The system performance of the communication system in which the communication device deploying the at least one AI model is located is lower than an eleventh threshold; The change in the neural network input and output distribution of the at least one AI model is greater than a twelfth threshold; The neural network weight distribution of the at least one AI model is greater than a thirteenth threshold; The change in the neural network weight of the at least one AI model is lower than the fourteenth threshold. The device according to any one of claims 47 to 56, characterized in that The processor is further configured to receive the first information; or Obtain measurement parameters of the first AI model and/or the second AI model, and determine the first information based on the measurement parameters of the first AI model. The device according to any one of claims 47 to 57, characterized in that the at least one AI model includes the first AI model and/or the second AI model; wherein the first AI model is associated with the second AI model, and the second AI model is used to be deployed on the second communication device. A first communication device, characterized by comprising a processor and a communication port; The processor is configured to obtain first information indicating performance monitoring information of at least one artificial intelligence (AI) model, where the at least one AI model is associated with a second AI model deployed on a second communication device; The communication port is used to send second information based on the first information, where the second information is obtained by performing noise disturbance processing based on the first parameter, and the second information is used to update the second AI model. The device according to claim 59 is characterized in that the second information is obtained by processing the quantization result of the first parameter, or the second information is obtained by processing the transmission parameter of the first parameter. The device according to claim 59 or 60, wherein the first parameter includes one or more of the following: The updated model parameters of the second AI model, used to update the gradient information of the second AI model, used to update the intermediate feature information of the second AI model, used to update the intermediate gradient information of the second AI model, the inference result of the first AI model, the distillation loss information corresponding to the second AI model, and the task loss information of the second AI model. The device according to any one of claims 59 to 61, characterized in that The communication port is also used for: When a first condition is met, sending second information based on the first information; The first condition includes at least one of the following: The training accuracy of the at least one AI model is lower than a first threshold; The test accuracy of the at least one AI model is lower than a second threshold; The system performance of the AI model system in which the at least one AI model is located is lower than a third threshold; The system performance of the communication system where the communication device deploying the at least one AI model is located is lower than a fourth threshold; The change in the neural network input and output distribution of the at least one AI model is greater than a fifth threshold; The neural network weight distribution of the at least one AI model is greater than a sixth threshold; The change in the neural network weight of the at least one AI model is lower than a seventh threshold. The device according to any one of claims 59 to 62, characterized in that The communication port is further used to receive indication information indicating the second parameter, and the second parameter is used to process the first parameter to obtain the second information. The device according to claim 63, characterized in that The communication port is further used to send request information for requesting the second parameter. The device according to any one of claims 59 to 64, characterized in that The processor is further configured to receive the first information; or Obtain measurement parameters of the first AI model and/or the second AI model, and determine the first information based on the measurement parameters of the first AI model. The device according to any one of claims 59 to 65 is characterized in that the at least one AI model includes the first AI model and/or the second AI model; wherein the first AI model is associated with the second AI model, and the second AI model is used to be deployed on the second communication device. A second communication device, characterized by comprising a processor and a communication port; The communication port is configured to receive second information, the second information being obtained by performing noise perturbation processing based on the first parameter; wherein the second information is used to update a second artificial intelligence (AI) model deployed in the second communication device; The processor is configured to update the second AI model based on the second information. The device according to claim 67 is characterized in that the second information is obtained by processing the quantization result of the first parameter, or the second information is obtained by processing the transmission parameter of the first parameter. The device according to claim 67 or 68, wherein the first parameter includes one or more of the following: The updated model parameters of the second AI model, used to update the gradient information of the second AI model, used to update the intermediate feature information of the second AI model, used to update the intermediate gradient information of the second AI model, the inference result of the first AI model, the distillation loss information corresponding to the second AI model, and the task loss information of the second AI model. The device according to any one of claims 67 to 69 is characterized in that the device is a chip or a chip system. A readable storage medium, characterized in that a computer program or instruction is stored in the storage medium, and when the computer program or instruction is executed by a computer, the method according to any one of claims 1 to 23 is implemented. A computer program product, characterized by comprising instructions, which, when the instructions are executed on a computer, cause the computer to perform the method according to any one of claims 1 to 23.