Communication method and related equipment

Abstract

The invention discloses a communication method and related equipment. The method comprises the following steps: generating a first artificial intelligence (AI) model; and sending first information to a second communication device, wherein the first information is used for determining the first AI model. After the first communication device generates the first AI model, the first communication device sends the first information used for determining the first AI model to the receiver so as to deploy the first AI model to the receiver, and the functions of deployment of the AI model, operation, generation or updating of the model and the like are achieved.

Description

Technical Field

[0001] This application relates to the field of communication technologies, and in particular, to a communication method and related devices. Background Art

[0002] Wireless communication can be a transmission communication between two or more communication nodes without propagation via a conductor or cable. Generally, such communication nodes include network devices and terminal devices.

[0003] Currently, with the development of communication technologies and the increasing maturity of artificial intelligence (AI) technologies, AI has gradually become an indispensable part of communication systems in wireless communication systems, and future wireless network architectures need to support a large number of AI functions. Therefore, how to support the deployment of models in wireless networks is an urgent problem to be solved.

[0004] In relevant standards, there are solutions for the combination of wireless networks and AI. However, the above solutions are all in the initial discussion stage of the combination of wireless networks and AI, and are not sufficient to support the deployment of AI. Summary of the Invention

[0005] This application provides a communication method and related devices, which can implement the deployment of an AI model and the operation of the model.

[0006] In a first aspect of this application, a communication method is provided. This method is executed by a first communication device. The first communication device can be a communication device (such as a network device), or the first communication device can be some components in a communication device (such as a processor, a chip, or a chip system, etc.), or the first communication device can also be a logical module or software that can implement all or part of the functions of a communication device. In this method, the first communication device generates a first artificial intelligence (AI) model and sends a first piece of information, and this first piece of information is used to determine the first AI model.

[0007] Based on the above technical solution, after the first communication device generates or updates the first AI model, by sending the first piece of information for determining the first AI model to the receiving party, the first AI model is deployed to the receiving party, realizing functions such as the deployment of the AI model, the operation of the model, generation, or update.

[0008] Optionally, "update" can be replaced with other terms, such as "modify", "iterate", "optimize", "process", or "receive from other devices", etc.

[0009] In this application, terms such as "AI model", "neural network model", "AI neural network model", "machine learning model", and "AI processing model" can be replaced with each other.

[0010] It should be understood that the first AI model group includes a first AI model and a second AI model, and it can be understood that the function of the first AI model group is implemented at least through the model processing of the first AI model and the model processing of the second AI model. In other words, after the first communication device receives the second information, the first communication device can deploy the first AI model to the first communication device through the model parameters of the first AI model group or the model parameters of the first AI model contained in the second information, and perform model processing on the first AI model; accordingly, the second communication device can perform model processing on the second AI model deployed on the second communication device. Optionally, the model processing may include one or more of model update processing, model training processing, and model inference processing.

[0011] It can be understood that the second communication device can be implemented in many ways.

[0012] For example, the second communication device may be a terminal device, and accordingly, the first communication device and the second communication device may communicate on a sidelink (SL). In this case, the first AI model and the second AI model may be referred to as an end-to-end model, or an end-to-end collaborative model, etc.

[0013] For another example, the second communication device may be a network device (e.g., an access network device), and accordingly, the first communication device and the second communication device may communicate on uplink and downlink communication links. In this case, the first AI model and the second AI model may be referred to as an edge-end model, an edge-end collaborative model, an edge-end model, an edge-end collaborative model, etc.

[0014] Optionally, when the first AI model group is regarded as one AI model, the first AI model and the second AI model may be understood as two AI sub-models in the one AI model.

[0015] In the present application, an AI model is deployed on a communication device (for example, the first AI model is deployed on the first communication device, the second AI model is deployed on the second communication device, etc.). It can be understood that after the communication device obtains the model parameters of the AI ​​model, it obtains / generates / constructs the AI ​​model based on the model parameters of the AI ​​model, and subsequently the communication device can perform model processing on the AI ​​model.

[0016] Optionally, the model parameters may include one or more of the model's hyperparameters, the model's data set (including the model's input data and label data corresponding to the input data), and the model's structural parameters.

[0017] Optionally, an AI model group may include two or more AI models. For example, in addition to the first AI model and the second AI model, the first AI model group may further include other AI models, which may be deployed on other communication devices different from the first communication device and the second communication device, and are not limited herein.

[0018] It should be understood that wireless communication signals (such as the transceiver of configuration information of communication resources, the transceiver of reference signals, etc.) may be transmitted between different communication devices (such as the first communication device and the second communication device).

[0019] Optionally, the AI models involved in this application (such as the first AI model, the second AI model, etc.) may be used to manage the wireless communication signals (including at least one of configuration, update, and optimization). For example, the AI model may include one or more of an AI model for modulation and / or demodulation, an AI model for channel prediction, an AI model for beam management, an AI model for assisted positioning, an AI model for channel compression, an AI model for resource scheduling, and an AI model for replacing one or more modules in a transmitter and / or a receiver. Alternatively, the AI models involved in this application may also be AI models for other AI tasks, such as an AI model for image recognition, an AI model for natural language processing, an AI model for computer vision, etc.

[0020] In a possible implementation manner of the first aspect, the first communication device is a functional entity for generating the first AI model.

[0021] Based on the above technical solution, the first communication device may be used to generate / acquire / determine / update one or more AI models.

[0022] In a possible implementation manner of the first aspect, the first communication device is a functional entity for generating the first AI model and deploying the first AI model to the second communication device through the first information.

[0023] Based on the above technical solution, the first communication device may communicate with one or more second communication devices, and the first communication device may send information of one or more first communication devices (such as one or more first information) to the second communication device to deploy the first AI model in one or more second communication devices.

[0024] In a possible implementation manner of the first aspect, the first AI model is a dedicated AI model.

[0025] Based on the above technical solution, the second communication device can be a terminal device. Therefore, the AI model deployed on this terminal device can be a dedicated AI model, and the first information sent by the first communication device to the second communication device can be used to determine this dedicated AI model. Since different terminal devices may have different end-side characteristics (such as different local data, different local computing power, different channel characteristics, etc.), by deploying a dedicated AI model in the terminal device, the AI model deployed in the terminal device can be adapted to the end-side characteristics of the terminal device, in order to improve the model processing performance of the AI model. In addition, compared with large models, deploying a dedicated AI model in the second communication device can effectively improve the model accuracy in personalized scenarios.

[0026] It should be understood that the general AI model can be referred to as the base model, large model or L0 model. The dedicated AI model can be called a small model, L1 model, L2 model, etc.

[0027] Taking the large model as an example, the large model can refer to a machine learning model with a large number of parameters and a complex structure, which can process massive data and complete various complex tasks, such as natural language processing, computer vision, speech recognition, etc.

[0028] Optionally, the large model is usually constructed by a deep neural network and has billions or even hundreds of billions of parameters.

[0029] Optionally, the design purpose of the large model can be to improve the expression ability and prediction performance of the model, and it can handle more complex tasks and data.

[0030] Optionally, the large model can learn complex patterns and features by training massive data, has a stronger generalization ability, and can make accurate predictions on data that has not been processed.

[0031] In contrast, the small model can refer to a model with fewer parameters and shallower layers. Generally, compared with the small model, the large model usually has more parameters and deeper layers, has a stronger expression ability and higher accuracy, but also requires more computing resources and time for training and inference, and is suitable for scenarios with a large amount of data and sufficient computing resources, such as cloud computing, high-performance computing, artificial intelligence, etc.

[0032] Optionally, the small model has the advantages of being lightweight, high-efficiency, easy to deploy, etc., and is suitable for scenarios with a small amount of data and limited computing resources, such as mobile applications, embedded devices, Internet of Things, etc.

[0033] In a possible implementation of the first aspect, before generating the first AI model, the first communication device includes: the first communication device receives second information from the second communication device; the first communication device generates the first AI model, including: the first communication device generates the first AI model according to the second information.

[0034] Based on the above technical solution, the first communication device can receive information (such as second information) from one or more second communication devices, and generate / acquire / determine / update one or more dedicated AI models according to the information of the one or more second communication devices. Optionally, the information of the second communication device can be used to indicate user demand information, user data information, channel state information, etc., so that the first communication device generates / acquires / determines / updates a dedicated AI model based on the information of the second communication device, thereby effectively improving the model accuracy in personalized scenarios.

[0035] In a possible implementation of the first aspect, the second information includes at least one of the following: the channel state information between the first communication device and the second communication device, and the local computing power state of the second communication device.

[0036] Based on the above technical solution, when the communication bandwidth between the first communication device and the second communication device is fixed, the first communication device can generate / acquire / determine / update one or more dedicated AI models according to the channel state information between the first communication device and the second communication device and / or the local computing power state of the second communication device, so as to generate a dedicated AI model in combination with the actual scenario requirements and improve the model accuracy in personalized scenarios.

[0037] Optionally, the channel state information may include the channel information between the first communication device and the second communication device, and / or the channel information between the second communication device and the first communication device. Among them, when the first communication device is a terminal device and the second communication device is a network device, the channel information between the first communication device and the second communication device can be understood as uplink channel information, and the channel information between the second communication device and the first communication device can be understood as downlink channel information.

[0038] Optionally, the channel state information between the first communication device and the second communication device can be obtained based on a reference signal.

[0039] For example, when the first communication device and the second communication device communicate through uplink and downlink links, the reference signal may include a channel state information reference signal (CSI-RS), a sounding reference signal (SRS), etc.

[0040] For another example, when the first communication device and the second communication device communicate via a sidelink, the reference signal may include a sidelink-synchronization signal / physical broadcast channel block (sidelink SSB, SL-SSB, or S-SS / PSBCH block), a sidelink-channel state information reference signal (SL-CSI-RS), and the like.

[0041] In a possible implementation of the first aspect, the second information may further include at least one of the following: the location information of the second communication device, the behavior information of the second communication device, the local data information of the second communication device, and the tag information.

[0042] Based on the above technical solution, the second information may include at least one of the above. In other words, the first communication device may determine the first AI model group based on at least one of the above information to improve the flexibility of the solution implementation.

[0043] In a possible implementation of the first aspect, the method further includes: the first communication device receives third information from a third communication device; the first communication device sends the third information to the second communication device, and the third information is used to determine a third AI model, and the third AI model is a general AI model.

[0044] Based on the above technical solution, since the general AI model is mainly used to ensure the basic requirements of model accuracy in most scenarios, and the dedicated AI model is mainly used to ensure the model accuracy in personalized scenarios, before or after deploying the dedicated AI model, the second communication device deploys the general AI model based on the third information of the first communication device, which can realize switching the dedicated AI model deployed on the second communication device to the general AI model, or switching the general AI model to the dedicated AI model, so as to realize the collaborative deployment of the dedicated AI model and the general AI model. That is, the first communication device can generate or update a matching AI model by combining the information fed back by itself, the second communication device, or other communication devices, and deploy the AI model that meets the scenario requirements on itself, the second communication device, or other communication devices in combination with the actual scenario, realizing the joint deployment of the dedicated AI model and the general AI model, effectively reducing the application cost of the model, and improving the model accuracy in personalized scenarios.

[0045] In a possible implementation of the first aspect, after the first communication device sends the first information to the second communication device, the method further includes: the first communication device receives fourth information from the second communication device, and the fourth information is used to indicate one of the AI models in the list of one or more AI models.

[0046] Based on the above technical solution, on the basis that the second communication device deploys a dedicated AI model, the second communication device can send the fourth information to the first communication device to feedback the recommended AI model.

[0047] Optionally, the list of AI models may include one or more AI models, and each AI model may include two or more AI models. As described above, the relationship between the AI model group and the AI model can also be understood as the relationship between the AI model and the AI sub-model. Therefore, the list of AI models can also be replaced by the list of AI model groups, that is, the list of AI model groups can include one or more AI models.

[0048] Optionally, the list can be replaced by other terms, such as set, dictionary, combination, space, etc.

[0049] In a possible implementation of the first aspect, the list of one or more AI models includes the list of one or more dedicated AI models and / or the list of general AI models.

[0050] Based on the above technical solution, the list of one or more AI models contains the list information of different AI models, so that the first communication device and / or the second communication device can select a matching AI model according to the actual needs to meet the accuracy requirements of different scenarios.

[0051] In a possible implementation of the first aspect, the first AI model is a general AI model.

[0052] Based on the above technical solution, the first AI model can be either a dedicated AI model or a general AI model to adapt to the accuracy requirements of different scenarios.

[0053] In a possible implementation of the first aspect, the first communication device generates the first AI model, including: the first communication device receives fifth information from the third communication device, and the fifth information is used to determine the first AI model, and the first AI model is deployed on the first communication device and the second communication device, or the first AI model is deployed on the second communication device.

[0054] Based on the above technical solution, the first communication device can generate or update the first AI model based on information from one or more third communication devices (such as one or more fifth information). Subsequently, multiple first communication devices and second communication devices connected to each first communication device can all deploy a general AI model with relatively high generalization and good versatility.

[0055] In a possible implementation manner of the first aspect, before the second communication device receives the first information from the first communication device, the method further includes: the first communication device sends a first message to the second communication device, and the first message is used to inform the second communication device to use the first AI model.

[0056] Based on the above technical solution, after the first communication device receives the fifth information, it can generate or update the first AI model based on the fifth information. Before the first communication device sends the first information to the second communication device, the first communication device can send a first message to the second communication device to inform the second communication device that it can use the first AI model (general AI model). It should be understood that the first message can also be used to inform the second communication device not to use the first AI model.

[0057] Optionally, the first message can be a broadcast, unicast or multicast message.

[0058] The first communication device sends the first message to one or more second communication devices in a broadcast, unicast or multicast manner, which can enable one or more second communication devices to obtain the first message through broadcast, unicast or multicast, thereby saving signaling overhead.

[0059] A second aspect of the present application provides a communication method, which is executed by the second communication device. The second communication device can be a communication device (such as a network device or a terminal device), or the second communication device can be part of the components in the communication device (such as a processor, a chip or a chip system, etc.), or the second communication device can also be a logical module or software that can implement all or part of the communication device functions. In this method, the second communication device receives the first information from the first communication device; the second communication device determines the first AI model according to the first information.

[0060] Based on the above technical solution, after the second communication device receives the first information, the second communication device can determine the first AI model based on the first information. In other words, as the receiver of the first information, the second communication device can determine the first AI model based on the first information from the first communication device to deploy the first AI model, thereby supporting the deployment of the model and realizing the operation of the model.

[0061] In a possible implementation manner of the second aspect, the first AI model is a dedicated AI model.

[0062] In a possible implementation of the second aspect, before the second communication device receives the first information from the first communication device, the method further includes: the second communication device sends second information to the first communication device, and the second information is used to generate the first AI model.

[0063] In a possible implementation of the second aspect, the second information includes at least one of the following: channel state information between the first communication device and the second communication device, local computing power state of the second communication device.

[0064] In a possible implementation of the second aspect, the second information further includes at least one of the following: location information of the second communication device, behavior information of the second communication device, local data information and label information of the second communication device.

[0065] In a possible implementation of the second aspect, the method further includes: the second communication device receives third information from the first communication device, and the third information is used to determine the third AI model, and the third AI model is a general AI model.

[0066] In a possible implementation of the second aspect, after the second communication device determines the first AI model according to the first information, the method further includes: the second communication device sends fourth information to the first communication device, and the fourth information is used to indicate one of the AI models in a list of one or more AI models.

[0067] In a possible implementation of the second aspect, the list of one or more AI models includes a list of one or more dedicated AI models and / or a list of general AI models.

[0068] In a possible implementation of the second aspect, the first AI model is a general AI model.

[0069] In a possible implementation of the second aspect, the method further includes: the second communication device receives fifth information from the first communication device, and the fifth information is used to determine the first AI model, and the first AI model is deployed on the first communication device and the second communication device, or the first AI model is deployed on the second communication device.

[0070] In a possible implementation of the second aspect, before the second communication device receives the first information from the first communication device, the method further includes: the second communication device receives a first message from the first communication device, and the first message is used to inform the second communication device to use the first AI model.

[0071] Optionally, the first message may be a broadcast, unicast, or multicast message. For the descriptions of various possible implementation manners of the second aspect of the embodiments of this application, reference may be made to the descriptions of various possible implementation manners in the first aspect, and details are not described herein again.

[0072] In a possible implementation manner of the first aspect or the second aspect, the first information is used to determine the category to which the second communication device belongs.

[0073] Based on the above technical solution, after receiving the second information from one or more second communication devices, the first communication device may classify the one or more second communication devices based on the second information, and include the relevant information of the category in the first information to determine the AI model corresponding to each category according to the category. Subsequently, the first communication device will separately send the first information to the one or more second communication devices. After the one or more second communication devices receive the first information from the first communication device, each second communication device may determine the category to which the second communication device belongs according to the first information, and screen out the AI model corresponding to the category from the first information according to the category to be determined, so as to implement the deployment of the AI model and the operation of the model in the second communication device. In a possible implementation manner of the first aspect or the second aspect, the first AI model is included in the first AI model group, and the first AI model group further includes a second AI model. The first AI model is deployed on the first communication device, and the second AI model is deployed on the second communication device; wherein, the input of the second AI model includes the output of the first AI model, or the input of the first AI model includes the output of the second AI model.

[0074] Based on the above technical solution, the first communication device may be a network device, and the second communication device may be a terminal device. Since different terminal devices may have different end-side characteristics (such as different local data, different local computing power, different channel characteristics, etc.), and different network devices may have different edge-side characteristics, by deploying the second AI model in the terminal device and deploying the first AI model in the network device, the second AI model deployed in the terminal device can be adapted to the end-side characteristics of the terminal device, and the first AI model deployed in the network device can be adapted to the edge-side characteristics of the network device, so as to improve the model processing performance of the AI model.

[0075] It can be understood that there are various implementations of the second communication device.

[0076] For example, the second communication device may be a terminal device. Correspondingly, the first communication device and the second communication device may communicate on a sidelink (SL). In this case, the first AI model and the second AI model may be referred to as an end-to-end model, or an end-to-end collaborative model, etc.

[0077] For another example, the second communication device may be a network device (such as an access network device). Correspondingly, the first communication device and the second communication device may communicate on the uplink and downlink communication links. In this case, the first AI model and the second AI model may be referred to as edge models, edge collaboration models, end-edge models, end-edge collaboration models, etc.

[0078] The cloud can be understood as the central node of traditional cloud computing and is the control end of edge computing. The edge can be understood as the edge side of cloud computing, which is divided into infrastructure edge and device edge, and refers to edge servers or base stations in a wireless network. The end can be understood as terminal devices, such as various terminals like mobile phones, tablets, sensors, etc.

[0079] In a possible implementation manner of the first aspect or the second aspect, the first information is further used to determine the second AI model in the first AI model group.

[0080] Based on the above technical solution, on the premise that the first AI model is deployed on the first communication device and the second AI model is deployed on the second communication device, the second communication device determines the AI models deployed on different devices by receiving the first information from the first communication device.

[0081] In a possible implementation manner of the first aspect or the second aspect, the first information includes the identifier of the first AI model, and the identifier is used to determine the first AI model in a list containing one or more dedicated AI models.

[0082] Based on the above technical solution, the first communication device may send the identifier of the first AI model to the second communication device to determine the first AI model deployed on the second communication device, thereby reducing the data transmission volume between the first communication device and the second communication device.

[0083] In a possible implementation manner of the first aspect or the second aspect, before sending the first information to the second communication device, the method further includes: the first communication device sends a list of one or more dedicated AI models to the second communication device.

[0084] Based on the above technical solution, the premise for the first communication device to send the identifier of the first AI model is that the second communication device has already stored a list of one or more dedicated AI models. The first communication device sends a list of one or more dedicated AI models to the second communication device in advance so that when the model type needs to be changed subsequently, the identifier of the model can be sent, thereby reducing the data transmission volume.

[0085] In a possible implementation manner of the first aspect or the second aspect, the first information includes the model parameters and / or model structure information of the first AI model.

[0086] In a possible implementation of the first aspect or the second aspect, the first information further includes an identifier of the first AI model.

[0087] Based on the above technical solution, on the basis that the first information includes the model parameters and / or model structure information of the first AI model, it may further include an identifier of the first AI model, so that the second communication device can determine the deployed AI model according to the identifier.

[0088] In a possible implementation of the first aspect or the second aspect, the first information is information sent periodically.

[0089] Based on the above technical solution, the first information can be one of the bases for determining the AI model. By periodically sending the first information between the first communication device and the second communication device, the periodic determination of the AI model can be achieved, so as to realize multiple iterative updates of the AI model through a periodic process.

[0090] The third aspect of the present application provides a communication method, which is executed by a third communication device. The third communication device may be a network device (such as an access network device, a core network device, a cloud server, etc.), or the third communication device may be a part of the components in the network device (such as a processor, a chip, or a chip system, etc.), or the third communication device may also be a logical module or software that can implement all or part of the communication device functions. In this method, the third communication device receives the sixth information from the first communication device; the third communication device generates the fifth information according to the sixth information; the third communication device sends the fifth information to the first communication device, and the fifth information is used to determine a general AI model, and the general AI model is deployed on the first communication device and the second communication device, or the general AI model is deployed on the second communication device.

[0091] Based on the above technical solution, the third communication device can generate a general AI model based on the sixth information from the first communication device and send the fifth information to the first communication device to deploy the general AI model on the first communication device. In addition, the general AI model can also be deployed on the second communication device, so as to realize the functions of supporting the deployment of the general AI model and running the model at both the first communication device and the second communication device.

[0092] Optionally, the sixth information includes the second information.

[0093] Optionally, the sixth information is determined based on the processed second information.

[0094] In a possible implementation of the third aspect, the sixth information includes at least one of the following:

[0095] Channel state information between the first communication device and the second communication device, local computing power state of the second communication device.

[0096] In a possible implementation of the third aspect, the sixth information further includes at least one of the following:

[0097] The location information of the second communication device, the behavior information of the second communication device, the local data information and label information of the second communication device.

[0098] For the descriptions of various possible implementations of the third aspect of this application embodiment, reference can be made to the descriptions of various possible implementations in the first aspect and the second aspect, which will not be elaborated here one by one.

[0099] The fourth aspect of this application provides a communication device, which is the first communication device. The device includes a processing unit and a transceiver unit; the processing unit is used to generate a first artificial intelligence (AI) model; the transceiver unit is used to send the first information, and the first information is used to determine the first AI model.

[0100] In the fourth aspect of this application, the component modules of the communication device can also be used to execute the steps performed in various possible implementations of the first aspect and achieve the corresponding technical effects. Specifically, reference can be made to the first aspect, which will not be elaborated here.

[0101] The fifth aspect of this application provides a communication device, which is the second communication device. The device includes a transceiver unit and a processing unit. The transceiver unit is used to receive the first information from the first communication device; the processing unit is used to determine the first AI model according to the first information.

[0102] In the fifth aspect of this application, the component modules of the communication device can also be used to execute the steps performed in various possible implementations of the second aspect and achieve the corresponding technical effects. Specifically, reference can be made to the second aspect, which will not be elaborated here.

[0103] The sixth aspect of this application provides a communication device, which is the third communication device. The device includes a transceiver unit and a processing unit; the transceiver unit is used to receive the sixth information from the first communication device; the processing unit is used to generate the fifth information according to the sixth information; the transceiver unit is used to send the fifth information to the first communication device, and the fifth information is used to determine the general AI model. The general AI model is deployed on the first communication device and the second communication device, or the general AI model is deployed on the second communication device.

[0104] In the sixth aspect of this application, the component modules of the communication device can also be used to execute the steps performed in various possible implementations of the third aspect and achieve the corresponding technical effects. Specifically, reference can be made to the third aspect, which will not be elaborated here.

[0105] The seventh aspect of the present application provides a communication device, including at least one processor, and the at least one processor is coupled to a memory; the memory is used to store programs or instructions; the at least one processor is used to execute the programs or instructions, so that the device implements the method of any possible implementation manner in any one of the foregoing first aspect to third aspect.

[0106] In a possible implementation manner, the communication device further includes a memory. Optionally, the processor and the memory are integrated together.

[0107] The eighth aspect of the present application provides a communication device, including at least one logic circuit and an input-output interface; the logic circuit is used to execute the method of any possible implementation manner in any one of the foregoing first aspect to third aspect.

[0108] The ninth aspect of the present application provides a communication system, which includes the foregoing first communication device and a second communication device. Alternatively, the communication system includes the foregoing first communication device, a second communication device, and a third communication device.

[0109] The tenth aspect of the present application 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 of any possible implementation manner in any one of the foregoing first aspect to third aspect.

[0110] The eleventh aspect of the present application provides a computer program product (or computer program). When the computer program in the computer program product is executed by the processor, the processor executes the method of any possible implementation manner in any one of the foregoing first aspect to third aspect.

[0111] The twelfth aspect of the present application provides a chip system, which includes at least one processor and is used to support a communication device to implement the method of any possible implementation manner in any one of the foregoing first aspect to third aspect.

[0112] In a possible design, the chip system may further include a memory, and the memory is used to store necessary program instructions and data of the communication device. The chip system may be composed of chips or may include chips and other discrete devices. Optionally, the chip system further includes an interface circuit, and the interface circuit provides program instructions and / or data for at least one processor.

[0113] Among them, for the technical effects brought by any one of the design manners in the fourth aspect to the twelfth aspect, reference may be made to the technical effects brought by different design manners in the foregoing first aspect to the third aspect, and details are not described herein again. Description of the Drawings

[0114] Figures 1a to 1d Schematic diagram of the communication system provided for this application;

[0115] Figures 2a to 2g Schematic diagram of the AI processing process involved in this application;

[0116] Figure 3 Schematic diagram of an implementation of the communication method provided by an embodiment of this application;

[0117] Figure 4 Another schematic diagram of the implementation of the communication method provided by an embodiment of this application;

[0118] Figures 5a to 5b Interaction schematic diagram of the communication method provided for this application;

[0119] Figure 6a Schematic diagram of the collaborative deployment of a dedicated AI model and a general AI model provided by an embodiment of this application;

[0120] Figures 6b to 6d Another schematic diagram of the implementation of the communication method provided by an embodiment of this application;

[0121] Figures 7 to 11 Schematic diagram of the communication device provided for this application. Detailed implementation manners

[0122] First, some terms in the embodiments of this application are explained to facilitate the understanding of those skilled in the art.

[0123] (1) Terminal device: It can be a wireless terminal device that can receive scheduling and indication information from a network device. The wireless terminal device can be a device that provides voice and / or data connectivity to users, or a handheld device with wireless connection functions, or other processing devices connected to a wireless modem.

[0124] A terminal device can communicate with one or more core networks or the Internet via a radio access network (RAN). The terminal device can be a mobile terminal device, such as a mobile phone (or a "cellular" phone, a mobile phone), a computer, and a data card. For example, it can be a portable, pocket-sized, handheld, computer-integrated, or vehicle-mounted mobile device that exchanges voice and / or data with the radio access network. For example, devices such as personal communication service (PCS) phones, cordless phones, session initiation protocol (SIP) phones, wireless local loop (WLL) stations, personal digital assistants (PDAs), tablets (Pads), and computers with wireless transceiver functions. The wireless terminal device can also be referred to as a system, a subscriber unit, a subscriber station, a mobile station, a mobile station (MS), a remote station, an access point (AP), a remote terminal device, an access terminal device, a user terminal device, a user agent, a subscriber station (SS), a customer premises equipment (CPE), a terminal, a user equipment (UE), a mobile terminal (MT), etc.

[0125] By way of example and not limitation, in the embodiments of the present application, the terminal device may also be a wearable device. A wearable device may also be referred to as a wearable intelligent device or a smart wearable device, etc. It is a general term for devices developed by applying wearable technology to the intelligent design of daily wear, such as glasses, gloves, watches, clothing, shoes, etc. A wearable device is a portable device that is worn directly on the body or integrated into the user's clothing or accessories. A wearable device is not only a hardware device, but also realizes powerful functions through software support, data interaction, and cloud interaction. Broadly speaking, wearable intelligent devices include those with complete functions and large sizes that can realize complete or partial functions without relying on a smart phone, such as smart watches or smart glasses, etc., and those that only focus on a certain type of application function and need to cooperate with other devices such as smart phones, such as various smart bracelets for vital sign monitoring, smart helmets, smart jewelry, etc.

[0126] The terminal may also be a drone, a robot, a terminal in device-to-device (D2D) communication, a terminal in vehicle to everything (V2X) communication, 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 remote medical, a wireless terminal in smart grid, a wireless terminal in transportation safety, a wireless terminal in smart city, a wireless terminal in smart home, etc.

[0127] In addition, the terminal device may also be a terminal device in a communication system evolved after the fifth generation (5G) communication system (such as the sixth generation (6G) communication system, etc.) or a terminal device in a future evolved public land mobile network (PLMN). Exemplarily, the 6G network can further expand the form and function of 5G communication terminals. 6G terminals include, but are not limited to, vehicles, cellular network terminals (integrating satellite terminal functions), drones, and Internet of Things (IoT) devices.

[0128] In the embodiments of the present application, the above terminal device may also obtain AI services provided by a network device. Optionally, the terminal device may also have AI processing capabilities.

[0129] (2) Network device: It can be a device in a wireless network. For example, the network device can be a RAN node (or device) that connects a terminal device to a wireless network, and can also be called a base station. Currently, some examples of RAN devices are: base station, evolved NodeB (eNodeB), gNB (gNodeB) in a 5G communication system, transmission reception point (TRP), evolved Node B (eNB), radio network controller (RNC), Node B (NB), home base station (e.g., home evolved Node B, or home Node B, HNB), base band unit (BBU), or wireless fidelity (Wi-Fi) access point AP, etc. Additionally, in a network structure, the network device can include a centralized unit (CU) node, or a distributed unit (DU) node, or a RAN device including a CU node and a DU node.

[0130] Optionally, the RAN node can also be a macro base station, a micro base station or an indoor station, a relay node or a donor node, or a radio controller in a cloud radio access network (CRAN) scenario. The RAN node can also be a server, a wearable device, a vehicle or an in-vehicle device, etc. For example, the access network device in vehicle to everything (V2X) technology can be a road side unit (RSU).

[0131] In another possible scenario, multiple RAN nodes cooperate to assist a terminal in achieving wireless access, and different RAN nodes respectively implement some functions of a base station. For example, the RAN node can be a central unit (CU), a distributed unit (DU), a CU-control plane (CP), a CU-user plane (UP), or a radio unit (RU), etc. The CU and DU can be separately provided, or can also be included in the same network element, such as a baseband unit (BBU). The RU can be included in a radio frequency device or a radio frequency unit, such as included in a remote radio unit (RRU), an active antenna unit (AAU), or a remote radio head (RRH).

[0132] In different systems, the CU (or CU-CP and CU-UP), DU, or RU may also have different names, but those skilled in the art can understand their meanings. For example, in an open RAN (O-RAN or ORAN) system, the CU can also be called an O-CU (open CU), the DU can also be called an O-DU, the CU-CP can also be called an O-CU-CP, the CU-UP can also be called an O-CU-UP, and the RU can also be called an O-RU. For the convenience of description, in this application, the CU, CU-CP, CU-UP, DU, and RU are used as examples for description. Any one of 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.

[0133] The communication between the access network device and the terminal device follows a certain protocol layer structure. The protocol layer may include a control plane protocol layer and a user plane protocol layer. The control plane protocol layer may include at least one of the following: Radio Resource Control (RRC) layer, Packet Data Convergence Protocol (PDCP) layer, Radio Link Control (RLC) layer, Media Access Control (MAC) layer, or Physical (PHY) layer, etc. The user plane protocol layer may include at least one of the following: Service Data Adaptation Protocol (SDAP) layer, PDCP layer, RLC layer, MAC layer, or Physical layer, etc.

[0134] For the correspondence between the network elements in the ORAN system and their realizable protocol layer functions, reference can be made to Table 1 below.

[0135] Table 1

[0136] ORAN network element Protocol layer functions of 3GPP O-CU-CP RRC+PCDP - Control Plane (PDCP-C) O-CU-UP SDAP+PCDP - User Plane (PDCP-U) O-DU RLC+MAC+PHY-high O-RU PHY-low

[0137] The network device may be other devices that provide wireless communication functions for the terminal device. The specific technologies and specific device forms adopted by the network device are not limited in the embodiments of the present application. For ease of description, the embodiments of the present application do not limit.

[0138] The network device may further include core network devices, 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, P-GW) in the 4th generation (4G) network; network elements such as 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 device may further include other core network devices in the 5G network and the next-generation network of the 5G network.

[0139] In the embodiments of this application, the above network device may also be 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, computing power node, RAN node with AI capabilities, or core network element with AI capabilities on the network side (access network or core network).

[0140] In the embodiments of this application, the device for implementing the functions of the network device may be the network device itself or a device capable of supporting the network device to implement such functions, such as a chip system, and this device may be installed in the network device. In the technical solutions provided in the embodiments of this application, the case where the device for implementing the functions of the network device is the network device itself is taken as an example to describe the technical solutions provided in the embodiments of this application.

[0141] (3) Configuration and pre-configuration: In this application, both configuration and pre-configuration are used. Among them, configuration means that the network device / server sends the configuration information or value of some parameters to the terminal through messages or signaling, so that the terminal can determine the communication parameters or resources during transmission according to these values or information. Pre-configuration is similar to configuration, and it can be the parameter information or value pre-negotiated between the network device / server and the terminal device, or the parameter information or value adopted by the base station / network device or terminal device specified by the standard protocol, or the parameter information or value pre-stored in the base station / server or terminal device. This application does not make any limitations in this regard.

[0142] Furthermore, these values and parameters can be changed or updated.

[0143] (4) The terms "system" and "network" in the embodiments of the present application may be used interchangeably. "Plurality" means two or more. "And / or" describes the association relationship of associated objects and indicates that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone, where A and B can be singular or plural. The character " / " generally represents an "or" relationship between the associated objects before and after. "At least one (item)" or its similar expression refers to any combination of these items, including any combination of single item (item) or plural items (items). For example, "at least one of A, B, and C" includes A, B, C, AB, AC, BC, or ABC. Also, 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, time sequence, priority, or importance degree of multiple objects.

[0144] (5) "Sending" and "receiving" in the embodiments of the present application represent the direction of signal transmission. For example, "sending information to XX" can be understood as the destination of the information is XX, which may include directly sending through the air interface and also include indirectly sending by other units or modules through the air interface. "Receiving information from YY" can be understood as the source of the information is YY, which may include directly receiving from YY through the air interface and also include indirectly receiving from YY through the air interface by 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.

[0145] In other words, sending and receiving can be carried out between devices, for example, between a network device and a terminal device, or can be carried out within a device, for example, sending or receiving between components, modules, chips, software modules, or hardware modules within a device through a bus, trace, or interface.

[0146] It can be understood that necessary processing may be performed on the information between the source end and the destination end of the information sending, such as encoding, modulation, etc., but the destination end can understand the valid information from the source end. Similar expressions in the present application can be understood similarly and will not be elaborated here.

[0147] (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. If the information indicated by a certain piece of information (such as the indication information described below) is called the information to be indicated, then in the specific implementation process, there are many ways to indicate the information to be indicated. For example, but not limited to, the information to be indicated can be directly indicated, such as the information to be indicated itself or the index of the information to be indicated, etc. It is also possible to indirectly indicate the information to be indicated by indicating other information, where there is an association relationship between the other information and the information to be indicated; it is also possible to only indicate a part of the information to be indicated, while the other parts of the information to be indicated are known or pre-agreed. For example, the arrangement order of each piece of information pre-agreed (such as protocol predefined) can be used to implement the indication of specific information, thereby reducing the indication overhead to a certain extent. The present application does not limit the specific manner of indication. It can be understood 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.

[0148] In the present application, unless otherwise specified, the same or similar parts between various embodiments can be referred to each other. In each embodiment of the present application, as well as in each method / design / implementation manner in each embodiment, if there is no special description and logical conflict, the terms and / or descriptions between different embodiments, as well as between each method / design / implementation manner in each embodiment, are consistent and can be referenced to each other. The technical features in different embodiments, as well as in each method / design / implementation manner in each embodiment, can be combined to form new embodiments, methods, or implementation manners according to their internal logical relationships. The embodiments of the present application described below do not constitute a limitation on the protection scope of the present application.

[0149] The present application can be applied to a long term evolution (LTE) system, a new radio (NR) system, or a communication system evolved after 5G (such as Beyond 5G (B5G), 6G, etc.). Among them, the communication system includes at least one network device and / or at least one terminal device.

[0150] Please refer to Figure 1a , which is a schematic diagram of the architecture of the communication system 1000 to which the embodiments of the present application are applied. As Figure 1a shown, the communication system includes a radio access network (RAN) 100 and a core network 200. Optionally, the communication system 1000 may also include the Internet 300. Among them, RAN 100 includes at least one RAN node (such as Figure 1a110a and 110b in (collectively referred to as 110) may also include at least one terminal (such as Figure 1a 120a - 120j in (collectively referred to as 120). RAN100 may also include other RAN nodes, for example, wireless relay devices and / or wireless backhaul devices ( Figure 1a not shown in the figure). The terminal 120 is connected to the RAN node 110 wirelessly, and the RAN node 110 is connected to the core network 200 wirelessly or by wire. The core network devices in the core network 200 and the RAN nodes 110 in the RAN100 may be independent different physical devices, or may be the same physical device integrating the logical functions of the core network devices and the logical functions of the RAN nodes. Terminals can be connected to each other and RAN nodes can be connected to each other either by wire or wirelessly.

[0151] RAN100 may be an evolved universal terrestrial radio access (E-UTRA) system, a new radio (NR) system, and a future radio access system defined in the 3rd generation partnership project (3GPP). RAN100 may also include two or more different radio access systems as described above. RAN100 may also be an open RAN (O-RAN).

[0152] For ease of description, in the following text, a base station is used as an example of a RAN node for description.

[0153] The base station and the terminal may be in a fixed position or movable. The base station and the terminal may be deployed on land, including indoor or outdoor, handheld or vehicle-mounted; may also be deployed on water; may also be deployed on airplanes, balloons, and artificial satellites. The embodiments of the present application do not limit the application scenarios of the base station and the terminal.

[0154] The roles of the base station and the terminal may be relative. For example, Figure 1a the helicopter or drone 120i in the figure may be configured as a mobile base station. For those terminals 120j accessing the radio access network 100 through 120i, the terminal 120i is a base station; but for the base station 110a, 120i is a terminal, that is, communication between 110a and 120i is through a wireless air interface protocol. Of course, communication between 110a and 120i may also be through an interface protocol between base stations. In this case, relative to 110a, 120i is also a base station. Therefore, both the base station and the terminal can be collectively referred to as communication devices. Figure 1a110a and 110b in it can be called communication devices with base station functions. Figure 1a 120a - 120j in it can be called communication devices with terminal functions.

[0155] Communication can be carried out between the base station and the terminal, between the base station and the base station, and between the terminal and the terminal through licensed spectrum, through unlicensed spectrum, or simultaneously through licensed spectrum and unlicensed spectrum; communication can be carried out through spectrum below 6 gigahertz (GHz), through spectrum above 6 GHz, or simultaneously using spectrum below 6 GHz and spectrum above 6 GHz. The embodiments of this application do not limit the spectrum resources used for wireless communication.

[0156] In the embodiments of this application, the functions of the base station can also be executed by modules (such as chips) in the base station, or by a control subsystem containing base station functions. The control subsystem containing base station functions here can be a control center in the above application scenarios such as smart grid, industrial control, intelligent transportation, and smart city. The functions of the terminal can also be executed by modules (such as chips or modems) in the terminal, or by a device containing terminal functions.

[0157] Figure 1b Another schematic diagram of a communication system provided for the embodiments of this application. In Figure 1b it, taking the network device as the base station as an example for illustration, both device 1 and device 2 are terminal devices. As Figure 1b shown, the communication link between device 1 and device 2 can be called a sidelink (SL), and the communication link between device 1 (or device 2) and the base station can be called an uplink and downlink link, including an uplink link and a downlink link; it can be seen that the sidelink is a communication mechanism in which different terminal devices communicate directly without passing through a network device.

[0158] Optionally, in the sidelink (SL), generally, the transmitting device and the receiving device can be terminal devices or network devices of the same type, or a roadside unit (RSU) and a terminal device. Here, from a physical entity perspective, the RSU is a roadside station or a roadside unit, and from a functional perspective, the RSU can be a terminal device or a network device. This application does not limit this. That is, the transmitting device is a terminal device and the receiving device is also a terminal device; or, the transmitting device is a roadside station and the receiving device is also a terminal device; or, the transmitting device is a terminal device and the receiving device is a roadside station. Additionally, the sidelink can also be base station devices of the same type or different types. In this case, the function of the sidelink is similar to that of a relay link, but the radio access technology used can be the same or different.

[0159] Exemplarily, broadcasting, unicasting, and multicasting are supported on the sidelink.

[0160] Broadcast communication is similar to a network device broadcasting system information, that is, the terminal device sends broadcast service data without encryption. Any other terminal device within the effective reception range can receive the data of this broadcast service if it is interested in this broadcast service.

[0161] Unicast communication is similar to the data communication after establishing an RRC connection between a terminal device and a network device. A unicast connection needs to be established between two terminal devices first. After establishing the unicast connection, the two terminal devices can perform data communication based on the negotiated identifier. This data can be encrypted or not encrypted. Compared with broadcasting, in unicast communication, only the two terminal devices that have established the unicast connection can perform this unicast communication.

[0162] Optionally, one unicast communication on the sidelink corresponds to a pair of source layer-2 identifiers (denoted as source L2 ID) and destination layer-2 identifiers (denoted as destination L2 ID). Optionally, the source L2 ID and the destination L2 ID will be included in the sub-header of the media access control protocol data unit (MAC PDU) in the sidelink to enable the data to be transmitted to the correct receiving end.

[0163] Multicast communication refers to the communication between all terminal devices within a communication group. Any terminal device within the group can send and receive the data of this multicast service.

[0164] Such as Figure 1cAs shown in the figure, when a terminal device (denoted as UE1) communicates directly with another terminal device (denoted as UE2) without passing through a network device, the communication link between the two terminal devices can be referred to as a sidelink, or it can be said that the two terminal devices communicate based on the proximity-based services communication 5 (PC5) interface.

[0165] As Figure 1d shown in the figure, as a typical application of sidelink, V2X communication technology utilizes and enhances the current cellular network functions and elements to achieve low-latency and high-reliability communication among various nodes in a vehicle network, including vehicle-to-vehicle communication (abbreviated as V2V), vehicle-to-pedestrian communication (abbreviated as V2P), vehicle-to-infrastructure communication (abbreviated as V2I), and vehicle-to-network communication (abbreviated as V2N). With the evolution of cellular systems from 4G Long Term Evolution (LTE) to 5G, C-V2X evolves from LTE-V2X to NR-V2X (New Radio V2X, abbreviated as NR-V2X).

[0166] In addition, V2X communication has great potential in reducing vehicle collision accidents, and thus can also reduce the corresponding number of casualties. The advantages of V2X are not limited to improving safety. Vehicles capable of V2X communication contribute to better traffic management, further promoting green transportation and lower energy consumption. The Intelligent Transportation System (ITS) is an application that combines with V2X. Based on V2X technology, Vehicle UEs (V-UEs) can send some of their own information, such as location, speed, intentions (turning, lane changing, reversing), etc., periodically and information triggered by some non-periodic events to surrounding V-UEs. Similarly, V-UEs will also receive information from surrounding users in real time. 5G NR V2X can support lower transmission latency, more reliable communication transmission, higher throughput, better user experience, and meet the requirements of a wider range of application scenarios. Further, the vehicle-to-vehicle communication technology supported by V2X can be extended to device-to-device (D2D) communication under any system.

[0167] The technical solution provided by this application can be applied to a wireless communication system (such as Figure 1a, Figure 1b , Figure 1c or Figure 1d The system shown), for example, an AI network element can be introduced in the communication system provided in this application to implement some or all of the 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 built into the network element of the communication system. For example, the AI network element can be an AI module built into: a terminal device, an access network device, a core network device, a cloud server, or a network management (operation, administration and maintenance, OAM) to implement AI-related functions. The OAM can be used as the network management of the core network device and / or as the network management of the access network device. Alternatively, the AI network element can also be an independently set network element in the communication system. Optionally, an AI entity can also be included in the terminal or the chip built into the terminal to implement AI-related functions.

[0168] The following briefly introduces the artificial intelligence (AI) that may be involved in this application.

[0169] Artificial intelligence (AI) can enable a machine to have human intelligence. For example, it can enable a machine to apply computer software and hardware to simulate certain intelligent behaviors of humans. To achieve artificial intelligence, machine learning methods can be used. In machine learning methods, a machine learns (or trains) a model using training data. The model represents the mapping from input to output. The learned model can be used for inference (or prediction), that is, the model can be used to predict the output corresponding to a given input. Among them, the output can also be referred to as an inference result (or prediction result).

[0170] Machine learning can include supervised learning, unsupervised learning, and reinforcement learning. Among them, unsupervised learning can also be referred to as non-supervised learning.

[0171] Supervised learning is based on the collected sample values and sample labels, uses machine learning algorithms to learn the mapping relationship from sample values to sample labels, and uses an AI model to express the learned mapping relationship. The process of training a machine learning model is the process of learning this mapping relationship. During the training process, the sample values are input into the model to obtain the predicted values of the model, and the model parameters are optimized by calculating the error between the predicted values of the model and the sample labels (ideal values). After the mapping relationship learning is completed, the learned mapping can be used to predict new sample labels. The mapping relationship learned by supervised learning can include linear mapping or non-linear mapping. According to the type of labels, the learning tasks can be divided into classification tasks and regression tasks.

[0172] Unsupervised learning discovers the intrinsic patterns of samples by itself using algorithms based on the collected sample values. In unsupervised learning, there is a type of algorithm that uses the samples themselves as the supervision signal, that is, the model learns the mapping relationship from samples to samples, which is called self-supervised learning. During training, the model parameters are optimized by calculating the error between the predicted value of the model and the samples themselves. Self-supervised learning can be used in applications such as signal compression and decompression recovery. Common algorithms include autoencoders and generative adversarial networks, etc.

[0173] Reinforcement learning is different from supervised learning. It is a type of algorithm that learns strategies to solve problems by interacting with the environment. Different from supervised and unsupervised learning, there is no clear "correct" action label data in reinforcement learning problems. The algorithm needs to interact with the environment to obtain the reward signal feedback from the environment, and then adjust the decision-making actions to obtain a larger reward signal value. For example, in downlink power control, the reinforcement learning model adjusts the downlink transmission power of each user according to the total system throughput rate feedback by the wireless network, and then expects to obtain a higher system throughput rate. The goal of reinforcement learning is also to learn the mapping relationship between the environmental state and the optimal (for example, the best) decision-making actions. However, because the labels of "correct actions" cannot be obtained in advance, the network cannot be optimized by calculating the error between the actions and the "correct actions". The training of reinforcement learning is achieved through iterative interaction with the environment.

[0174] Neural network (NN) is a specific model in machine learning technology. According to the universal approximation theorem, neural networks can theoretically approximate any continuous function, enabling neural networks to have the ability to learn any mapping. Traditional communication systems need to rely on rich expert knowledge to design communication modules, while deep learning communication systems based on neural networks can automatically discover implicit pattern structures from large datasets, establish the mapping relationship between data, and obtain better performance than traditional modeling methods.

[0175] The idea of neural networks comes from the neuron structure of the brain tissue. For example, each neuron performs a weighted sum operation on its input values and outputs the operation result through an activation function.

[0176] As Figure 2a shown, it is a schematic diagram of a neuron structure. Assume the input of the neuron is x = [x 0 , x 1 , …, x n , and the weights corresponding to each input are w = [w, w 1 , …, w n , where n is a positive integer, and w i and x i can be various possible types such as decimals, integers (such as 0, positive integers, or negative integers, etc.), or complex numbers. wi As x i The weight value is used to weight x i For weighted summation. The bias for weighted summation of the input value is, for example, b. There can be various forms of activation functions. Assuming the activation function of a neuron is: y = f(z) = max(0, z), then the output of this neuron is: For another example, the activation function of a neuron is: y = f(z) = z, then the output of this neuron is: Among them, b can be various possible types such as a decimal, an integer (such as 0, a positive integer, or a negative integer), or a complex number. The activation functions of different neurons in a neural network can be the same or different.

[0177] In addition, a neural network generally includes multiple layers, and each layer can include one or more neurons. By increasing the depth and / or width of the neural network, the expression ability of the neural network can be improved, providing a more powerful information extraction and abstract modeling ability for complex systems. Among them, the depth of the neural network can refer to the number of layers included in the neural network, and the number of neurons included in each layer can be called the width of that layer. In one implementation, a neural network includes an input layer and an output layer. The input layer of the neural network processes the received input information through neurons and passes the processing result to the output layer, and the output layer obtains the output result of the neural network. In another implementation, a neural network includes an input layer, a hidden layer, and an output layer. The input layer of the neural network processes the received input information through neurons and passes the processing result to the intermediate hidden layer. The hidden layer calculates the received processing result to obtain a calculation result, and the hidden layer passes the calculation result to the output layer or the next adjacent hidden layer, and finally the output layer obtains the output result of the neural network. Among them, a neural network can include one hidden layer, or include multiple successively connected hidden layers, without limitation.

[0178] A neural network is, for example, a deep neural network (DNN). According to the construction method of the network, DNN can include a feedforward neural network (FNN), a convolutional neural network (CNN), and a recurrent neural network (RNN).

[0179] Figure 2b It is a schematic diagram of an FNN network. The characteristic of the FNN network is that neurons between adjacent layers are completely connected in pairs. This characteristic makes FNN usually require a large amount of storage space and result in a high computational complexity.

[0180] A CNN is a neural network specifically designed to process data with a similar grid structure. For example, time series data (discretely sampled along the time axis) and image data (two-dimensionally discretely sampled) can both be considered data with a similar grid structure. Instead of using all the input information for computation at once, a CNN uses a window of a fixed size to intercept partial information for convolution operations, which greatly reduces the computational amount of model parameters. Additionally, depending on the type of information intercepted by the window (such as people and objects in the same picture being different types of information), different convolution kernels can be used for each window, enabling the CNN to better extract the features of the input data.

[0181] An RNN is a type of DNN network that utilizes feedback time series information. Its input includes the new input value at the current moment and its own output value at the previous moment. RNNs are suitable for capturing sequential features that are relevant over time and are particularly applicable to applications such as speech recognition and channel coding and decoding.

[0182] During the model training process of the above-mentioned machine learning, a loss function can be defined. The loss function describes the gap or difference between the output value of the model and the ideal target value. The loss function can be embodied in various forms, and no specific form of the loss function is restricted. The model training process can be regarded as the following process: by adjusting some or all of the parameters of the model, the value of the loss function is made less than the threshold value or meets the target requirements.

[0183] The model can also be referred to as an AI model, a rule, or other names, etc. An AI model can be considered as a specific method for implementing AI functions. An AI model represents the mapping relationship or function between the input and output of the model. AI functions can include one or more of the following: data collection, model training (or model learning), model information publishing, model inference (or also known as model reasoning, reasoning, or prediction, etc.), model monitoring or model verification, or the release of inference results, etc. AI functions can also be referred to as AI (related) operations or AI-related functions.

[0184] Next, the implementation process of the neural network will be described exemplarily in conjunction with the accompanying drawings.

[0185] 1. Fully connected neural network, also known as a multilayer perceptron (MLP).

[0186] As Figure 2c shown, an MLP contains an input layer (left side), an output layer (right side), and multiple hidden layers (in the middle). Among them, each layer of the MLP contains several nodes, called neurons. Among them, the neurons between adjacent layers are pairwise connected.

[0187] Optionally, considering neurons in adjacent two layers, the output h of the neurons in the next layer is the weighted sum of all the neurons x in the previous layer connected to it and passes through an activation function, which can be expressed as:

[0188] h = f(wx + b).

[0189] Among them, w is the weight matrix, b is the bias vector, and f is the activation function.

[0190] Further optionally, the output of the neural network can be recursively expressed as:

[0191] y = f n (w n f n-1 (…)+b n ).

[0192] Among them, n is the index of the neural network layer, 1 <= n <= N, where N is the total number of layers of the neural network.

[0193] In other words, the neural network can be understood as a mapping relationship from the input data set to the output data set. Usually, the neural network is randomly initialized, and the process of obtaining this mapping relationship from the existing data with random w and b is called the training of the neural network.

[0194] Optionally, the specific training method is to evaluate the output result of the neural network by using a loss function.

[0195] As Figure 2d shown, the error can be backpropagated, and the neural network parameters (including w and b) can be iteratively optimized by the method of gradient descent until the loss function reaches the minimum value, that is, Figure 2d the "preferred point (such as the optimal point)" in Figure 2d . It can be understood that the neural network parameters corresponding to the "preferred point (such as the optimal point)" in

[0196] Further optionally, the process of gradient descent can be expressed as:

[0197]

[0198] Among them, θ is the parameter to be optimized (including w and b), L is the loss function, η is the learning rate, which controls the step size of gradient descent, represents the derivative operation, represents the derivative of L with respect to θ.

[0199] Further optionally, the process of backpropagation makes use of the chain rule of partial derivatives.

[0200] AsFigure 2e As shown, the gradient of the previous layer parameters can be recursively calculated by the gradient of the next layer parameters, which can be expressed as:

[0201]

[0202] Among them, w ij is the weight of node j connecting node i, s i is the weighted sum of the inputs to node i.

[0203] 2. Federated Learning (FL)

[0204] 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 tasks by promoting the collaboration of various edge devices and central servers.

[0205] like Figure 2f As shown, the FL architecture is a training architecture in the current FL field. For example, the FedAvg algorithm is the basic algorithm of FL, and its algorithm flow is roughly as follows:

[0206] (1) The center initializes the model to be trained And broadcast it to all client devices.

[0207] (2) In 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.

[0208] (3) The central node aggregates and collects 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.

[0209] (4) Repeat steps (2) and (3) until the model finally converges or the number of training rounds reaches the upper limit.

[0210] 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.

[0211] As can be seen, in the FL framework, the dataset exists at distributed nodes, that is, the distributed nodes collect the local dataset and perform local training, and report the local results (models or gradients) obtained from the training to the central node. The central node itself does not have a dataset and is only responsible for fusing the training results of the distributed nodes to obtain a global model and distributing it to the distributed nodes.

[0212] 3. Decentralized learning. Different from federated learning, another distributed learning architecture - decentralized learning.

[0213] As Figure 2g shown, consider a fully distributed system without a central node. The design objective f(x) of the decentralized learning system is generally the mean of the objective functions f i (x) of each node, that is where n is the number of distributed nodes, x is the parameter to be optimized, and in machine learning, x is the parameter of the machine learning (such as neural network) model. Each node uses local data and local objective f i (x) to calculate the local gradient and then sends it to the neighbor nodes that are communication-reachable. 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:

[0214]

[0215] where represents the parameter of the local model after the (k + 1)th (k is a natural number) update in the i-th node, represents the parameter of the local model after the k-th update in the i-th node (if k is 0, it means is the parameter of the local model of the i-th node that has not participated in the update), α k represents the tuning coefficient, N i is the set of neighbor nodes of node i, and |N i | represents the number of elements in the set of neighbor nodes of node i, that is, the number of neighbor nodes of node i. Through the information interaction between nodes, the decentralized learning system will finally learn a unified model.

[0216] The technical solution provided by this application can be applied to a wireless communication system (such as Figure 1a or Figure 1bIn the system shown, with the development of communication technologies and the increasing maturity of artificial intelligence (AI) technologies, AI has gradually become an indispensable part of wireless communication systems. Future wireless network architectures need to support a large number of AI functions. Therefore, how to support the deployment of models in wireless networks is an urgent problem to be solved.

[0217] Future intelligent wireless networks need to be able to support a large number of AI / machine learning (ML) functions, including the deployment and update of general AI models and dedicated AI models. However, there is no mature and widely recognized technical solution to support how wireless networks support the deployment of general and dedicated AI models, and this gap urgently needs to be filled.

[0218] Currently, in relevant standards, there are solutions for the combination of wireless networks and AI. The first type of deployment solution is:

[0219] (1) AI / ML model training is located in OAM (operation, administrator and maintenance), and AI / ML model inference is located in gNB.

[0220] (2) Both AI / ML model training and AI / ML model inference are located in gNB.

[0221] For the first type of deployment, that is, model training is located in OAM and model inference is located in the next generation radio access network (NG-RAN).

[0222] If the 5G base station adopts a split architecture, that is, gNB can adopt a split architecture, and gNB is composed of a control unit (CU) and one or more distributed units (DUs). The interface between gNB CU and gNB DU is called F1. The second type of deployment solution is:

[0223] (1) AI / ML model training is located in OAM, and AI / ML model inference is located in gNB-CU.

[0224] (2) Both AI / ML model training and AI / ML model inference are located in gNB-CU.

[0225] For the second type of deployment, that is, both model training and model inference are located in NG-RAN.

[0226] It should be understood that in the future intelligent wireless network for 6G, there should be a scenario where a general AI model and a dedicated AI model coexist, and the wireless network architecture needs to support functions such as the deployment, generation, or update of the general AI model and the dedicated AI model. However, the above solutions (such as the first type and the second type of deployment solutions) are all in the initial discussion stage of the combination of wireless networks and AI. The radio air interface signaling for the purpose of supporting connections or sessions will not be sufficient to support the deployment of models, and the signaling process between model training and inference also fails to provide a deployment solution for the dedicated AI model and / or the general AI model.

[0227] To solve the above problems, an embodiment of the present application provides a communication method. Please refer to Figure 3 , Figure 3 which is a schematic implementation diagram of the communication method provided by the embodiment of the present application. The method includes the following steps.

[0228] It should be noted that in Figure 3 , the first communication device and the second communication device are used as the execution subjects of this interaction schematic to illustrate the method, but the present application does not limit the execution subjects of this interaction schematic. For example, in Figure 3 , the execution subject of the method can be replaced by a chip, a chip system, a processor, a logic module, or software in the communication device. In Figure 3 , the first communication device can be a network device and the second communication device can be a terminal device, or both the first communication device and the second communication device are terminal devices (for example, the method can be applied to the communication process of different terminal devices in the sidelink communication scenario).

[0229] S301. The first communication device generates a first AI model.

[0230] S302. The first communication device sends first information, and the first information is used to determine the first AI model. Correspondingly, the second communication device receives the first information.

[0231] S303. The second communication device determines the first AI model according to the first information.

[0232] In the present application, terms such as AI model, neural network model, AI neural network model, machine learning model, and AI processing model can be replaced with each other.

[0233] It should be understood that wireless communication signals (such as the transceiver of configuration information of communication resources, the transceiver of reference signals, etc.) can be transmitted between different communication devices (such as the first communication device and the second communication device).

[0234] Optionally, the AI models involved in this application (such as the first AI model, the second AI model, and the third AI model and the fourth AI model mentioned later, etc.) can be used to manage the wireless communication signal (including at least one of configuration, update, and optimization). For example, the AI model can include one or more of an AI model for modulation and / or demodulation, an AI model for channel prediction, an AI model for beam management, an AI model for assisting positioning, an AI model for channel compression, an AI model for resource scheduling, and an AI model for replacing one or more modules in a transmitter and / or a receiver. Alternatively, the AI model involved in this application can also be an AI model for other AI tasks, such as an AI model for image recognition, an AI model for natural language processing, an AI model for computer vision, etc.

[0235] Optionally, when the first AI model group is regarded as one AI model, the first AI model and the second AI model can be understood as two AI sub-models in this one AI model.

[0236] In this application, when an AI model is deployed on a communication device (for example, the first AI model is deployed on the first communication device, and the second AI model is deployed on the second communication device, etc.), it can be understood that after the communication device obtains the model parameters of the AI model, the communication device obtains / generates / constructs the AI model based on the model parameters of the AI model, and subsequently, the communication device can perform model processing on the AI model.

[0237] Optionally, the model parameters can include one or more of the hyperparameters of the model, the data set of the model (including the input data of the model and the label data corresponding to the input data), and the structure parameters of the model.

[0238] In a possible implementation, the first communication device is a functional entity for generating or updating the first AI model, that is, generating / obtaining / determining / updating one or more AI models.

[0239] In a possible implementation, the first communication device is a functional entity for generating or updating the first AI model and deploying the first AI model on the second communication device through the first information. Specifically, the first communication device can communicate with one or more second communication devices, and the first communication device can send information of one or more first communication devices (such as one or more first information) to the second communication device to deploy the first AI model on one or more second communication devices.

[0240] For the convenience of description, the first AI model is taken as an example of a dedicated AI model and a general AI model respectively for description.

[0241] It should be understood that a general AI model can be referred to as a base model, a large model, or an L0 model. A dedicated AI model can be referred to as a small model, an L1 model, an L2 model, etc.

[0242] Taking the large model as an example, a large model can refer to a machine learning model with a large number of parameters and a complex structure, capable of processing massive amounts of data and completing various complex tasks, such as natural language processing, computer vision, speech recognition, etc.

[0243] Optionally, a large model is usually constructed by a deep neural network and has billions or even hundreds of billions of parameters.

[0244] Optionally, the design purpose of a large model can be to improve the model's expressive ability and prediction performance, and it can handle more complex tasks and data.

[0245] Optionally, a large model can learn complex patterns and features by training on massive amounts of data, has a more powerful generalization ability, and can make accurate predictions on data that has not been processed.

[0246] In contrast, a small model can refer to a model with fewer parameters and shallower layers. Generally, compared with small models, large models usually have more parameters and deeper layers, have stronger expressive ability and higher accuracy, but also require more computing resources and time for training and inference, and are suitable for scenarios with large amounts of data and sufficient computing resources, such as cloud computing, high-performance computing, artificial intelligence, etc.

[0247] Optionally, small models have the advantages of being lightweight, highly efficient, and easy to deploy, and are suitable for scenarios with small amounts of data and limited computing resources, such as mobile applications, embedded devices, the Internet of Things, etc.

[0248] 1. In one possible implementation, the first AI model is a dedicated AI model.

[0249] Correspondingly, the first communication device generates a dedicated AI model in step S301, and the second communication device determines the dedicated AI model according to the first information in step S303. Specifically, the second communication device can be a terminal device, and the AI model deployed on the terminal device can be a dedicated AI model, and the first information sent by the first communication device to the second communication device can be used to determine the dedicated AI model. Since different terminal devices may have different end-side characteristics (such as different local data, different local computing power, different channel characteristics, etc.), by deploying a dedicated AI model in the terminal device, in this way, the AI model deployed on the terminal device can be adapted to the end-side characteristics of the terminal device, in order to improve the model processing performance of the AI model. In addition, compared with large models, deploying a dedicated AI model in the second communication device can effectively improve the model accuracy in personalized scenarios.

[0250] In a possible implementation, the first communication device generates a dedicated AI model in step S301, including: receiving second information from the second communication device; generating a dedicated AI model according to the second information. Specifically, the first communication device may receive second information from one or more second communication devices and generate a dedicated AI model according to the second information. For the sake of convenience of description, taking the example that the first communication device communicates with two second communication devices respectively, to Figure 4 specifically illustrate the deployment mechanism of the dedicated AI model. Please refer to Figure 4 , Figure 4 which is another schematic diagram of the implementation of the communication method provided by the embodiment of the present application, including the following steps:

[0251] S401. The first communication device receives second information from the first second communication device.

[0252] S402. The first communication device receives second information from the second second communication device.

[0253] After the first second communication device and the second communication device complete the initial access, they can respectively feedback the second information to the first communication device.

[0254] S403. The first communication device generates corresponding dedicated AI models according to the second information of the two second communication devices respectively.

[0255] S404. The first communication device sends first information to the first second communication device, and the first information is used to determine the dedicated AI model.

[0256] S405. The first communication device sends first information to the second second communication device, and the first information is used to determine the dedicated AI model.

[0257] In this step, after the first communication device generates the first information corresponding to the two second communication devices according to the second information, by sending the corresponding first information to one or more second communication devices respectively, the dedicated AI models are respectively deployed in the corresponding second communication devices.

[0258] Steps S404 and S405 can be understood as an implementation example of the foregoing step S302.

[0259] In a possible implementation, the first communication device generates a dedicated AI model according to the second information in step S403, including: the first communication device performs at least one of the processes of migrating, fine-tuning, distilling, and pruning on the general AI model.

[0260] Optionally, the large model can learn complex patterns and features by training on a vast amount of data, and has a more powerful generalization ability, enabling it to make accurate predictions on unprocessed data. In contrast, a small model can refer to a model with fewer parameters and shallower layers. The first communication device can obtain a lightweight and parameter-sparse dedicated AI model by migrating, fine-tuning, distilling, and pruning a general AI model.

[0261] In a possible implementation, in step S402, the first communication device can classify the second communication device according to one or more pieces of second information, and generate a dedicated AI model corresponding to each type of the second communication device (such as a single model or a set of edge-side models). Specifically, the first communication device can classify the second communication device according to the second information, and add a type identifier (such as: type 1, type 2, etc.) to each dedicated AI model, thereby generating the first information, that is, a list of dedicated AI models.

[0262] Optionally, the list can be replaced with other terms, such as set, dictionary, combination, space, etc.

[0263] Optionally, the first communication device can generate a list of dedicated AI models based on the information bottleneck (IB) theory.

[0264] In a possible implementation, the second information can include one or more of the following information A to information E.

[0265] Information A. Channel state information between the first communication device and the second communication device.

[0266] Information B. Local computing power status of the second communication device.

[0267] Information C. Location information of the second communication device.

[0268] Information D. Behavioral information of the second communication device.

[0269] Information E. Local data information and label information of the second communication device.

[0270] For information A, optionally, the channel state information between the first communication device and the second communication device can be obtained based on reference signals.

[0271] For example, when the first communication device and the second communication device communicate via uplink and downlink, the reference signal may include a channel state information reference signal (CSI-RS), a sounding reference signal (SRS), etc.

[0272] For another example, when the first communication device and the second communication device communicate via sidelink, the reference signal may include a sidelink-synchronization signal / physical broadcast channel block (sidelink SSB, SL-SSB, or S-SS / PSBCH block), a sidelink-channel state information reference signal (SL-CSI-RS), etc.

[0273] Optionally, the channel state information may include the channel information between the first communication device and the second communication device, and / or the channel information between the second communication device and the first communication device. Among them, when the first communication device is a terminal device and the second communication device is a network device, the channel information between the first communication device and the second communication device can be understood as uplink channel information, and the channel information between the second communication device and the first communication device can be understood as downlink channel information.

[0274] Regarding information B, when the second information includes the local computing power status information of the second communication device, since the complexity requirements of the model processing of the AI model may be related to the local computing power status of the second communication device. Therefore, for the first communication device, the dedicated AI model determined based on the local computing power status information can be adapted to the local computing power status of the second communication device, so as to provide the second communication device with a dedicated AI model that meets the local computing power status, and improve the processing performance of the first communication device for model processing based on the dedicated AI model.

[0275] Regarding information C, when the second information includes the location information of the second communication device, since the category to which the second communication device belongs may change when the location of the second communication device changes. Therefore, for the first communication device, the dedicated AI model determined based on the location information can be adapted to the location of the second communication device to generate a dedicated AI model that meets the accuracy requirements.

[0276] For information D, when the second information includes the behavior information of the second communication device, since the parameters or structure of the dedicated AI model may change when the behavior of the second communication device changes. Therefore, for the first communication device, the dedicated AI model determined by the first communication device based on the location information is used to generate a dedicated AI model that meets the accuracy requirements.

[0277] Optionally, the behavior information includes one or more of the user's movement trajectory, RRC state, and access or handover-related behaviors.

[0278] For information E, when the second information includes the local data information and label information of the second communication device, to ensure the data security of the second communication device or prevent privacy leakage, for the AI tasks where the second communication device needs to upload the original local data, the data can be encrypted, such as homomorphic encryption, etc. In the AI tasks where the second communication device does not need to upload the original local data, the second communication device can send the local data to the first communication device in a compressed manner, in an embedding or other non-one-to-one mapping form, to ensure the security of the data.

[0279] In a possible implementation manner, the second communication device can determine the category to which the second communication device belongs according to the first information in step S303. Specifically, the first information includes the category to which the second communication device belongs, and the first communication device sends the first information to the second communication device so that the second communication device determines the category to which it belongs according to the first information.

[0280] It should be understood that as the second communication device moves, the channel and location of the second communication device will change, and the category to which it belongs may change. However, since the second communication device periodically reports the second information to the first communication device, the first communication device will periodically update the category to which the second communication device belongs, and send the type identifier and dedicated AI model of this category to the second communication device. Since the category of the second communication device has changed, the first communication device needs to send a new model category (and / or) model to the second communication device, so Figure 4 The described solution is applicable to scenarios where the second communication device is relatively fixed or moves slowly. In this scenario, the frequency of the first communication device sending the model is low, so the overhead is small.

[0281] Next, the content of the list of one or more AI models included in the first information will be described. Among them, the content of the first information is related to the first AI model, where the first AI model is a single AI model or the first AI model is included in the first AI model group.

[0282] (1) The first AI model is a single AI model

[0283] As described above, the first information can be used to determine the category to which the second communication device belongs. In one possible implementation, the first information includes an identifier of a first AI model, and the identifier type is used to determine the first AI model in a list containing one or more dedicated AI models. Exemplarily, the content of the list of dedicated AI models is shown in Table 2 below.

[0284] Table 2

[0285] type model type 1 Dedicated AI model

[0286] In one possible implementation, the first information includes an identifier, that is, the first communication device sends the identifier type to the second communication device.

[0287] Optionally, before the first communication device sends the first information to the second communication device, it sends a list of one or more dedicated AI models to the second communication device. Specifically, the first communication device sends a list of one or more dedicated AI models to the second communication device in advance, so that when the type of the model needs to be changed subsequently, the identifier type of the model can be sent, thereby reducing the data transmission volume. The second communication device can determine the AI model corresponding to the identifier from Table 2 by looking up the table or other means.

[0288] It should be understood that after the category of the second communication device changes, the situation where the first communication device sends the identifier of the category to which the second communication device belongs is more applicable to the scenario where the second communication device moves relatively fast. In this scenario, although the type of the second communication device changes relatively fast, since the first communication device needs to send the updated identifier to the second communication device instead of sending the updated model, the signaling overhead is small.

[0289] Optionally, the list can be replaced with other terms, such as set, dictionary, combination, space, etc.

[0290] In another possible implementation, the first information includes the model parameters and / or model structure information of the first AI model, that is, the first communication device sends the model parameters and / or model structure information of the first AI model to the second communication device. Specifically, the second communication device can directly determine the deployed first AI model based on the model parameters and / or model structure information of the first AI model without performing the above-mentioned table lookup operation.

[0291] In this implementation mode, optionally, the first information further includes an identifier of the first AI model. Specifically, in addition to sending the model parameters and / or model structure information of the first AI model to the second communication device, the first communication device can also send the identifier of the first AI model to the second communication device, so that the second communication device can determine the first AI model based on the identifier or the parameters of the model, etc.

[0292] (2) The first AI model is included in the first AI model group.

[0293] In a possible implementation, the first AI model is included in the first AI model group, and the first AI model group further includes a second AI model. The first AI model is deployed on a first communication device, and the second AI model is deployed on a second communication device. Wherein, the input of the second AI model includes the output of the first AI model, or the input of the first AI model includes the output of the second AI model.

[0294] In this implementation, it is assumed that the first communication device can be a network device, and the second communication device can be a terminal device. Since different terminal devices may have different end-side characteristics (such as different local data, different local computing power, different channel characteristics, etc.), and different network devices may have different edge-side characteristics, by deploying the second AI model in the terminal device and the first AI model in the network device, it can enable the second AI model deployed in the terminal device to adapt to the end-side characteristics of the terminal device, and enable the first AI model deployed in the network device to adapt to the edge-side characteristics of the network device, in order to improve the model processing performance of the AI model.

[0295] Optionally, an AI model group can include two or more AI models. For example, in addition to including the first AI model and the second AI model, the first AI model group can also include other AI models, and these other AI models can be deployed on other communication devices different from the first communication device and the second communication device, which is not limited here.

[0296] Optionally, the AI model group list can include one or more AI model groups, and each AI model group can include two or more AI models. As mentioned above, the relationship between the AI model group and the AI model can also be understood as the relationship between the AI model and the AI sub-model. Therefore, the AI model group list can also be replaced by the AI model list, that is, the AI model list can include one or more AI models.

[0297] Optionally, the list can be replaced by other terms, such as set, dictionary, combination, space, etc.

[0298] Optionally, when the first AI model group is regarded as one AI model, the first AI model and the second AI model can be understood as two AI sub-models in the one AI model.

[0299] Optionally, the first AI model and the second AI model included in the first AI model group can be general AI models or dedicated AI models to achieve the update of different types of models.

[0300] In a possible implementation, the first information includes an identifier of the first AI model, and the identifier is used to determine the first AI model from a list including one or more dedicated AI models.

[0301] In this implementation, optionally, before sending the first information to the second communication device, the method further includes: sending a list of one or more dedicated AI models to the second communication device.

[0302] It should be understood that similar to the case where the first AI model is a single model, in this implementation, the content of the first information may include an identifier (type). Before receiving the first information, the second communication device stores in advance a list of one or more dedicated AI models. After the second communication device receives the first information from the first communication device, it can find the corresponding AI model from the list according to the identifier in the first information.

[0303] It should be understood that after the category of the second communication device changes, the situation where the first communication device sends the identifier of the category to which the second communication device belongs is more applicable to the scenario where the second communication device moves relatively fast. In this scenario, although the type of the second communication device changes relatively fast, since the first communication device needs to send the updated identifier to the second communication device instead of sending the updated model, the signaling overhead is relatively small.

[0304] In a possible implementation, the first information includes model parameters and / or model structure information of the first AI model.

[0305] In a possible implementation, the first information further includes an identifier (type) of the first AI model.

[0306] In the above two implementations, the content of the first information may be model parameters and / or model structure information of the first AI model, or the content of the first information is model parameters and / or model structure information of the first AI model and the corresponding identifier (type and AI model) of the first AI model.

[0307] In the case where the first AI model is included in the first AI model group, the first information sent by the first communication device to the second communication device may include information of most models in the first model group.

[0308] In another possible implementation, the first information is further used to determine a second AI model in the first AI model group.

[0309] Exemplarily, in the case where the first AI model is included in the first AI model group, the content of the first information may be as shown in Table 3 below.

[0310] Table 3

[0311] type Model 1 Model 2 type 1 First AI model Second AI model

[0312] Based on the content of Table 2 and Table 3, it should be understood that the content of the first information includes the following forms:

[0313] 1. Type (applicable to the case where a list has been saved. The type can be sent down so that the first communication device can find the corresponding AI model from the list according to the type, thereby reducing signaling overhead).

[0314] 2. Type and a single AI model.

[0315] 3. Type and a group of AI models.

[0316] 4. Type and the AI model on the terminal side.

[0317] For ease of understanding, the following will use the examples shown by Figure 5a and Figure 5b to give an example description of the first AI model deployed by the first communication device and the second communication device.

[0318] As shown in the example of Figure 5a , the first AI model is deployed on the first communication device, and the second AI model is deployed on the second communication device. Moreover, the input of the first AI model deployed on the first communication device includes the output of the second AI model deployed on the second communication device. In this example, taking the input data of the second AI model as X, after being processed by the second AI model, the second communication device can obtain and send data Z; after being transmitted through the wireless channel, the data received by the first communication device is represented as (It can be understood that due to transmission path loss and interference such as noise on the wireless channel, may not be the same as Z, which can be understood as an estimate of Z or a measured value of Z, etc.). Thereafter, the first communication device can use the data as the input of the first AI model, and after being processed by the first AI model, obtain data

[0319] As shown in the example of Figure 5b , the first AI model is deployed on the first communication device, and the second AI model is deployed on the second communication device. Moreover, the input of the second AI model includes the output of the first AI model. In this example, taking the input data of the first AI model as X, after being processed by the first AI model, the first communication device can obtain and send data Z; after being transmitted through the wireless channel, the data received by the second communication device is represented as Thereafter, the second communication device can use the data as the input of the second AI model, and after being processed by the second AI model, obtain data

[0320] It can be understood that the first communication device and the second communication device may be implemented in many ways.

[0321] For example, the second communication device may be a terminal device, and accordingly, the first communication device and the second communication device may communicate on a sidelink (SL). In this case, the first AI model and the second AI model may be referred to as an end-to-end model, or an end-to-end collaborative model, etc.

[0322] For another example, the second communication device may be a network device (e.g., an access network device), and accordingly, the first communication device and the second communication device may communicate on the uplink and downlink communication links. In this case, the first AI model and the second AI model may be referred to as an edge-end model, an edge-end collaborative model, an edge-end model, an edge-end collaborative model, etc. Exemplarily, when the first communication device is a terminal device and the second communication device is an access network device, Figure 5a The scenario shown can be understood as end-edge collaboration based on the downlink scenario. Figure 5b The scenario shown can be understood as end-edge collaboration implemented based on the uplink scenario.

[0323] It should be noted that in Figure 5a and Figure 5b In the example, data Y may be label data corresponding to data X, and the label data Y and the processing results of the first AI model and the second AI model are The association relationship between the first AI model and the second AI model can be used to detect or determine the processing performance of the first AI model and the second AI model. For example, the association relationship can be determined by gradient information, loss function, etc.

[0324] Based on the introduction of the dedicated AI model, the coordinated deployment of the dedicated AI model and the general AI model is explained below.

[0325] See also Figure 6a , Figure 6a A schematic diagram of the collaborative deployment of a dedicated AI model and a general AI model provided in an embodiment of the present application.

[0326] Assume that the first communication device is a base station and the second communication device is a terminal device, corresponding to Figure 6a , cloud can be understood as the central node of traditional cloud computing, which is the control end of edge computing. Cloud / core network can correspond to the third communication device described in this embodiment. Edge can be understood as the edge side of cloud computing, which is divided into infrastructure edge and device edge. In wireless network, it refers to edge server or base station, which can correspond to the first communication device of this embodiment. End can be understood as terminal equipment, such as mobile phones, tablets, sensors and other types of terminals, which can correspond to the second communication device of this embodiment.

[0327] likeFigure 6a As shown, the general AI model is mainly generated by the cloud or the core network, or pre-configured by the first communication device. The dedicated AI model is generated by the first communication device. Through the above implementation methods, the AI model collaboration on the terminal-edge-cloud side can be achieved.

[0328] In a possible implementation method, please refer to Figure 6b , Figure 6b which is another schematic diagram of the communication method provided by the embodiments of the present application. Figure 6b It is mainly used to introduce the deployment scheme of the general AI model. Figure 6b It is mainly used to illustrate that the user adopts the latest AI model sent by the base station, including the following steps:

[0329] S601. The first communication device receives the third information from the third communication device.

[0330] After the first communication device receives the third information from the third communication device, it can determine the general AI model according to the third information and store the general AI model.

[0331] Optionally, the general AI model can also be pre-configured locally by the first communication device without receiving the third information generated and sent by the cloud / core network.

[0332] S602. The first communication device sends the third information to the second communication device. The third information is used to determine the third AI model, and the third AI model is the general AI model.

[0333] S603. The second communication device determines the third AI model according to the third information.

[0334] S604. The second communication device determines the first AI model according to the first information.

[0335] In step S604, for the sake of brief description, only the content of the second communication device determining the first AI model according to the first information is shown. In actual execution, after step S603 is completed, the second communication device can feedback other requirements to the first communication device. The first communication device responds to this requirement and sends the first information corresponding to the requirement to the second communication device. Or, the first communication device sends the first information to the second communication device so that the second communication device determines the first AI model according to the first information and switches the deployed AI model from the third AI model to the first AI model.

[0336] Optionally, before Figure 6b the step S601 shown, the following can also be executed Figure 3Steps S301 to S303, that is, the second communication device may first deploy or store a dedicated AI model, and then deploy or store a general AI model, so as to deploy a matching AI model according to actual requirements, and implement the joint deployment of the dedicated AI model and the general AI model.

[0337] Optionally, after Figure 6b the step S604 shown, Figure 3 steps S301 to S303 may also be executed, that is, the second communication device may first deploy or store a general AI model, and then deploy or store a dedicated AI model, so as to deploy a matching AI model according to actual requirements, and implement the joint deployment of the dedicated AI model and the general AI model.

[0338] Exemplarily, please refer to Figure 6c , Figure 6c which is another schematic diagram of the implementation of the communication method provided by the embodiment of the present application. Taking the first communication device as a base station (BS) and the second communication device as a user equipment (UE) as an example for illustration. The steps are as follows:

[0339] 1. The base station sends the category of the general AI model (for example: type 0) and the end-side model (the UE has been informed through SIB that the general AI model needs to be used after access) to the UE.

[0340] 2. After the UE initially accesses, it uses the type 0 end-side model.

[0341] 3. The UE feeds back channel state information, etc. (i.e., the second information) to the base station.

[0342] 4. The base station classifies the UE and determines the category to which the UE belongs according to the list of dedicated AI models.

[0343] 5. The base station sends the corresponding category (for example: type 1) and the end-side dedicated AI model to the UE.

[0344] 6. The UE uses the latest type 1 end-side model.

[0345] It should be understood that for the sake of brief description, the specific source of the general AI model is omitted. The base station may receive the third information sent from the cloud / core network to determine the general AI model, or the base station obtains the general AI model according to the configuration information. Step 1 corresponds to Figure 6b step S602 in Figure 6b , step 2 corresponds to Figure 6b step S603 in

[0346] It should be understood that steps 3, 4, and 6 are all optional steps. For steps 3 and 4, the base station can send the category corresponding to the UE and the dedicated AI model on the terminal side to the UE according to the second information fed back by the UE, or can also send down the category corresponding to the UE and the dedicated AI model on the terminal side to the UE according to its own needs or the needs fed back by other communication devices. For step 6, after receiving the category corresponding to the UE and the dedicated AI model from the base station, the UE can choose to adopt and deploy the dedicated AI model, or can first store the category (for example: type 1) and the dedicated AI model corresponding to the category, and deploy it when there is a need for the dedicated AI model.

[0347] Optionally, after the base station sends the entire list of dedicated AI models to the UE, as the UE moves, the base station can notify the UE of the update of its category.

[0348] Optionally, the third communication device can generate a third AI model based on emulator pre-training or live network data, and send third information to the second communication device, where the third information is used to determine the third AI model.

[0349] Optionally, the third AI model is included in the second AI model group, and the second AI model group further includes a fourth AI model. The third AI model is deployed on the first communication device, and the fourth AI model is deployed on the second communication device; wherein, the input of the third AI model includes the output of the fourth AI model, or the input of the fourth AI model includes the output of the third AI model.

[0350] In step S601, in addition to obtaining the third information from the third communication device, optionally, the third information can also be default-borne by the first communication device. For example, the first communication device obtains the third information based on prior configuration (such as default configuration), stores the third information locally or in the cloud, and obtains it from the local or the cloud when the first communication device has a need to deploy a general AI model. After the first communication device receives the third information from the third communication device, the first communication device can receive the third information sent by the second communication device to determine the third AI model, that is, the general AI model, and deploy the general AI model on the second communication device.

[0351] In addition to obtaining the third information from the first communication device, optionally, the third information can also be default-borne by the second communication device. For example, the second communication device obtains the third information based on prior configuration (such as default configuration), stores the third information locally or in the cloud, and obtains it from the local or the cloud when the second communication device has a need to deploy a general AI model.

[0352] Optionally, the first communication device and / or the second communication device itself has or stores a general AI model (e.g., default configuration), and the general AI model can be stored locally or in the cloud of the first communication device and / or the second communication device.

[0353] Optionally, the third information can be understood as a list of general AI models. Taking the general AI model as a single model, the list of general AI models can be as shown in Table 4 below. The content and distribution method of the list of general AI models are similar to those of the list of dedicated AI models. For specific details, please refer to the relevant description of the list of dedicated AI models above and will not be elaborated here.

[0354] Table 4

[0355] type model type 0 General AI model

[0356] In this embodiment, the general AI model is represented by the type of type 0, and the dedicated AI model is represented by the type of type 1. In the actual execution process, the type corresponding to different AI models can be set according to specific requirements, which is only for illustrative purposes here and is not limited.

[0357] Optionally, after the second communication device determines the third AI model according to the third information in step S603, if the second communication device receives the first information from the first communication device, the second communication device determines the first AI model according to the first information and uses the first AI model, that is, the dedicated AI model, for communication or transmission.

[0358] In a possible implementation manner, after the first communication device sends the first information to the second communication device, the method further includes: the first communication device receives the fourth information from the second communication device, and the fourth information is used to indicate one of the AI models in the list of one or more AI models.

[0359] In this implementation manner, on the basis that the second communication device has deployed the dedicated AI model, the second communication device can actively send the fourth information to the first communication device to feedback the recommended AI model (e.g., type 3).

[0360] Specifically, it includes two categories:

[0361] The first category: The first communication device adopts the recommendation of the second communication device.

[0362] 1. The second communication device feeds back the model type currently recommended by the second communication device to the first communication device.

[0363] 2. The first communication device makes a decision and feeds back to the second communication device: Optionally, this type.

[0364] 3. The second communication device runs the end-side model of this type.

[0365] Category 2: The first communication device rejects the suggestion of the second communication device and provides the type selection of the first communication device.

[0366] The second communication device feeds back to the first communication device the model type currently suggested by the second communication device.

[0367] The first communication device decides not to recommend selecting this type and feeds back the type suggested by the UE (e.g., type 2).

[0368] The second communication device operates the end-side model of this type according to the instructions of the first communication device.

[0369] Exemplarily, please refer to Figure 6d , Figure 6d , which is another schematic diagram of the communication method provided by the embodiments of this application. Taking the first communication device as the base station (BS) and the second communication device as the user equipment (UE) as an example for illustration. Figure 6d It is mainly used to illustrate that the UE has the right to feed back the locally preferred model type to the base station, and the suggestion of the UE needs to interact with the decision of the base station.

[0370] 1. The base station sends the category of the general AI model (e.g., type 0) and the end-side model (it has been informed the UE through SIB that the general AI model needs to be used after access) to the UE.

[0371] 2. After the UE initially accesses, it uses the end-side model of type 0.

[0372] 3. The UE feeds back channel state information, etc. (i.e., the second information) to the base station.

[0373] 4. The base station generates a list of dedicated AI models.

[0374] 5. The base station sends the list of dedicated AI models to the UE (if it is an end-edge collaborative model group, it can choose to send only the category and the end-side model).

[0375] 6. The UE selects a model based on the list of dedicated AI models.

[0376] 7. The UE feeds back the suggested model selection (e.g., type 3) to the base station.

[0377] 8. The base station judges the feasibility of the model selection suggested by the UE.

[0378] 9. If the decision of the base station is feasible, it feeds back a message of model selection confirmation (ACK) to the UE, otherwise it feeds back a message of model selection rejection (NACK) to the UE and provides the selection of the base station (e.g., type 2).

[0379] 10. If the UE receives the confirmation information feedback from the base station, it uses the previously provided recommended selection type to the base station (e.g., type 3). If the UE receives the rejection information feedback from the base station, it uses the selection provided by the base station (e.g., type 2).

[0380] It should be understood that for the sake of brief description, the specific source of the general AI model is omitted. The base station can receive the third information sent from the cloud / core network to determine the general AI model, or the base station obtains the general AI model according to the configuration information. Step 1 corresponds to Figure 6b step S602 in, step 2 corresponds to Figure 6b step S603 in. When the UE uses the end-side model of type 0, step 5 corresponds to Figure 6b step S604 in.

[0381] It should be understood that steps 3, 4, 6 to 10 are all optional steps. For steps 3 and 4, the base station can send the category corresponding to the UE and the end-side dedicated AI model to the UE according to the second information feedback by the UE, or send the category corresponding to the UE and the end-side dedicated AI model to the UE according to its own needs or the needs feedback by other communication devices. For step 6, after receiving the category corresponding to the UE and the end-side dedicated AI model from the base station, the UE can choose to adopt and deploy the dedicated AI model, or can first store the category (e.g., type1) and the dedicated AI model corresponding to the category, and deploy it when there is a need for the dedicated AI model. For steps 7 to 10, whether the UE feedbacks the recommended model selection to the base station is determined according to the needs of the UE during actual execution, and it is an optional step. The UE can feedback the model selection recommended by the UE side (e.g., type 3) to the base station. If the base station's decision is feasible, the UE can adopt the recommended model selection (e.g., type3). If the base station rejects the recommendation, the UE adopts the model selection recommended by the base station (e.g., type 2).

[0382] It should be understood that since the base station has more local information, the priority of the base station's selection is higher than the UE's recommended selection.

[0383] Optionally, in this implementation manner, the list of one or more AI models includes the list of one or more dedicated AI models and / or the list of general AI models.

[0384] It should be understood that the list of one or more AI models can include the content of the dedicated AI model and / or the content of the general AI model, that is, it can be any content in Table 2, Table 3, Table 4 above, or the combined content shown in Table 2 and Table 4, Table 3 and Table 4 above, or Table 2, Table 3 and Table 4 above. Among them, it can be specifically set according to actual needs and is not limited here.

[0385] Second, in a possible implementation, the first AI model is a general AI model.

[0386] In a possible implementation, the general AI model is generated by a third communication device. Specifically, the third communication device receives sixth information from the first communication device, generates fifth information according to the sixth information, the third communication device sends the fifth information to the first communication device, and the fifth information is used to determine the general AI model. The general AI model is deployed on the first communication device and the second communication device, or the general AI model is deployed on the second communication device.

[0387] Based on the above technical solution, the third communication device generates a general AI model based on the sixth information from the first communication device, and sends the fifth information to the first communication device to deploy the general AI model on the first communication device. The first communication device can generate or update the general AI model through information from one or more third communication devices (such as one or more fifth information). Subsequently, multiple first communication devices and the second communication devices connected to each first communication device can all deploy a general AI model with relatively high generalization and good versatility.

[0388] Optionally, the sixth information includes the second information. Specifically, the first communication device receives the second information from the second communication device and sends the sixth information to the third communication device, where the sixth information includes all the content of the second information, or the sixth information includes part of the content of the second information. For example, only the data that the third communication device needs to use is sent, and other data is omitted to reduce the data transmission volume.

[0389] Optionally, the sixth information is determined based on the processed second information. Specifically, after the first communication device receives the second information from the second communication device, the first communication device processes the second information according to its own or AI requirements from the third communication device and sends the processed information (i.e., the sixth information) to the third communication device. Or, the first communication device extracts a list of general AI models from the second information and sends the list of general AI models to the third communication device, etc.

[0390] Optionally, the third communication device is the cloud side or the core network.

[0391] In a possible implementation, before the first communication device receives the first information from the first communication device, the method further includes: the first communication device sends a broadcast message to the second communication device, and the broadcast message is used to inform the second communication device to use the first AI model. Correspondingly, the second communication device receives the broadcast message from the first communication device.

[0392] It should be understood that the first communication device can inform the second communication device whether it needs to use the general AI model after its access by sending a broadcast message (such as a system information block (SIB)) to the second communication device. If so, after the second communication device accesses, the first communication device can send the first information to the second communication device. If not, the first communication device does not send the first information to the second communication device, but sends the first information based on a request from the second communication device when the second communication device has relevant requirements.

[0393] In a possible implementation manner, one or more of the above first information, second information, third information, fourth information, fifth information, and sixth information are information sent periodically.

[0394] Specifically, the first information can be one of the determination bases of the AI model, the second information can deploy a dedicated AI model, the third information can deploy a general AI model, the fourth information can provide model suggestions, and the fifth information can deploy a general AI model. Among them, by periodically sending the first information and / or the second information and / or the third information and / or the fourth information between the first communication device and the second communication device, the periodic determination and / or periodic deployment and / or periodic suggestions of the AI model can be realized, so as to realize multiple iterative updates of the AI model through a periodic process. By periodically sending the fifth information and / or the sixth information between the first communication device and the third communication device, the periodic deployment of the AI model can be realized, so as to realize multiple iterative updates of the AI model through a periodic process.

[0395] It should be understood that the specific implementation manner of the general AI model is similar to that of the dedicated AI model. For the specific implementation content of the general AI model, reference can be made to the description in the dedicated AI model, which will not be elaborated here.

[0396] Please refer to Figure 7 , an embodiment of the present application provides a communication device 700. The communication device 700 can implement the functions of the second communication device or the first communication device in the above method embodiment, and thus can also achieve the beneficial effects possessed by the above method embodiment. In the embodiment of the present application, the communication device 700 can be the first communication device (or the second communication device), or an integrated circuit or component inside the first communication device (or the second communication device), such as a chip.

[0397] It should be noted that the transceiver unit 702 can include a sending unit and a receiving unit, which are respectively used to perform sending and receiving.

[0398] In a possible implementation, when the device 700 is used to execute the method performed by the first communication device in the foregoing embodiment, the device 700 includes a processing unit 701 and a transceiver unit 702; the processing unit 701 is used to generate a first artificial intelligence (AI) model; the transceiver unit 702 is used to send first information, and the first information is used to determine the first AI model.

[0399] In a possible implementation, when the device 700 is used to execute the method performed by the second communication device in the foregoing embodiment, the device 700 includes a processing unit 701 and a transceiver unit 702; the transceiver unit 702 is used to receive the first information from the first communication device; the processing unit 701 is used to determine the first AI model according to the first information.

[0400] In a possible implementation, when the device 700 is used to execute the method performed by the third communication device in the foregoing embodiment, the device 700 includes a processing unit 701 and a transceiver unit 702; the transceiver unit 702 is used to receive sixth information from the first communication device; the processing unit 701 is used to generate fifth information according to the sixth information; the transceiver unit 702 is used to send the fifth information to the first communication device, and the fifth information is used to determine a general AI model, and the general AI model is deployed in the first communication device and the second communication device, or the general AI model is deployed in the second communication device.

[0401] It should be noted that for the content such as the information execution process of the units of the foregoing communication device 700, reference may be specifically made to the description in the method embodiment shown in the foregoing of this application, and details are not described herein again.

[0402] Please refer to Figure 8 , which is another schematic structural diagram of the communication device 800 provided by this application. The communication device 800 includes a logic circuit 801 and an input / output interface 802. Among them, the communication device 800 may be a chip or an integrated circuit.

[0403] Among them, Figure 7 the shown transceiver unit 702 may be a communication interface, and this communication interface may be Figure 8 the input / output interface 802 in , and the input / output interface 802 may include an input interface and an output interface. Alternatively, this communication interface may also be a transceiver circuit, and this transceiver circuit may include an input interface circuit and an output interface circuit.

[0404] Optionally, the logic circuit 801 is used to generate a first artificial intelligence (AI) model; the input / output interface 802 is used to send first information, and the first information is used to determine the first AI model.

[0405] Optionally, the input / output interface 802 is configured to receive first information from a first communication device; the logic circuit 801 is configured to determine a first AI model according to the first information.

[0406] Optionally, the input / output interface 802 is configured to receive sixth information from the first communication device; the logic circuit 801 is configured to generate fifth information according to the sixth information; the input / output interface 802 is configured to send the fifth information to the first communication device, and the fifth information is used to determine a general AI model, where the general AI model is deployed on the first communication device and the second communication device, or the general AI model is deployed on the second communication device.

[0407] Wherein, the logic circuit 801 and the input / output interface 802 may also perform other steps performed by the first communication device or the second communication device in any of the embodiments and achieve corresponding beneficial effects, which will not be elaborated here.

[0408] In a possible implementation manner, Figure 7 the shown processing unit 701 may be Figure 8 the logic circuit 801 in

[0409] Optionally, the logic circuit 801 may be a processing device, and the functions of the processing device may be implemented partially or fully by software. Among them, the functions of the processing device may be implemented partially or fully by software.

[0410] Optionally, the processing device may include a memory and a processor. 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 of the method embodiments.

[0411] Optionally, the processing device may include a processor. The memory for storing the computer program is located outside the processing device, and the processor is connected to the memory through a circuit / wire to read and execute the computer program stored in the memory. Among them, the memory and the processor may be integrated together, or may be physically independent of each other.

[0412] 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 processor units (CPUs), network processors (NPs), digital signal processing circuits (DSPs), microcontroller units (MCUs), programmable logic devices (PLDs), or other integrated chips, or any combination of the above chips or processors, etc.

[0413] Please refer to Figure 9 , the communication device 900 involved in the above embodiments provided by the embodiments of the present application. The communication device 900 may specifically be the communication device as the terminal device in the above embodiments. Figure 9 The example shown is implemented by the terminal device (or a component in the terminal device).

[0414] Among them, a possible schematic logical structure of the communication device 900. The communication device 900 may include but is not limited to at least one processor 901 and a communication port 902.

[0415] Among them, Figure 7 The shown transceiver unit 702 may be a communication interface. The communication interface may be Figure 9 the communication port 902 in

[0416] Further optionally, the device may further include at least one of a memory 903 and a bus 904. In the embodiments of the present application, the at least one processor 901 is used to control and process the actions of the communication device 900.

[0417] In addition, the processor 901 may be a central processing unit, a general-purpose processor, a digital signal processor, an application-specific integrated circuit, a field-programmable gate array, or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute various exemplary logic blocks, modules, and circuits described in connection 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 digital signal processor and a microprocessor, and so on. Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the systems, devices, and units described above can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0418] It should be noted that Figure 9 the communication device 900 shown can specifically be used to implement the steps implemented by the terminal device in the foregoing method embodiments and achieve the corresponding technical effects of the terminal device. Figure 9 For the specific implementation manners of the communication device shown, reference can be made to the descriptions in the foregoing method embodiments, and details will not be repeated here.

[0419] Please refer to Figure 10 , which is a schematic structural diagram of the communication device 1000 involved in the foregoing embodiments provided in the embodiments of the present application. The communication device 1000 can specifically be the communication device acting as a network device in the foregoing embodiments. Figure 10 The example shown is implemented by a network device (or a component in the network device). Among them, the structure of the communication device can refer to Figure 10 the structure shown.

[0420] The communication device 1000 includes at least one processor 1011 and at least one network interface 1014. Further optionally, the communication device further includes at least one memory 1012, at least one transceiver 1013, and one or more antennas 1015. The processor 1011, the memory 1012, the transceiver 1013, and the network interface 1014 are connected, for example, through a bus. In the embodiments of the present application, this connection may include various interfaces, transmission lines, or buses, etc., and this embodiment does not limit this. The antenna 1015 is connected to the transceiver 1013. The network interface 1014 is used to enable the communication device to communicate with other communication devices through a communication link. For example, the network interface 1014 may include a network interface between the communication device and a core network device, such as an S1 interface. 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.

[0421] Among them, Figure 7 the transceiver unit 702 shown may be a communication interface, and this communication interface may beFigure 10 The network interface 1014 in it, and the network interface 1014 may include an input interface and an output interface. Alternatively, the network interface 1014 may also be a transceiver circuit, and the transceiver circuit may include an input interface circuit and an output interface circuit.

[0422] The processor 1011 is mainly used to process communication protocols and communication data, and control the entire communication device, execute software programs, and process data of software programs. For example, it is used to support the communication device to execute the actions described in the embodiments. The communication device may include a baseband processor and a central processor. The baseband processor is mainly used to process communication protocols and communication data, and the central processor is mainly used to control the entire terminal device, execute software programs, and process data of software programs. Figure 10 The processor 1011 in it may integrate the functions of the baseband processor and the central processor. Those skilled in the art can understand that the baseband processor and the central processor may also be independent processors, interconnected through technologies such as a bus. Those skilled in the art can understand that the terminal device may include multiple baseband processors to adapt to different network modes, and the terminal device may include multiple central processors to enhance its processing ability. Each component of the terminal device may be connected through various buses. The baseband processor may also be expressed as a baseband processing circuit or a baseband processing chip. The central processor may also be expressed as a central processing circuit or a central processing chip. The function of processing communication protocols and communication data may be built into the processor or stored in the memory in the form of a software program, and the processor executes the software program to implement the baseband processing function.

[0423] The memory is mainly used to store software programs and data. The memory 1012 may exist independently and be connected to the processor 1011. Optionally, the memory 1012 may be integrated with the processor 1011, for example, integrated within a single chip. Among them, the memory 1012 can store the program code for implementing the technical solution of the embodiments of the present application and be controlled by the processor 1011 to execute. Various computer program codes executed can also be regarded as the driver programs of the processor 1011.

[0424] Figure 10 Only one memory and one processor are shown. 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, etc. The memory may be a storage element on the same chip as the processor, that is, an on-chip storage element, or an independent storage element. The embodiments of the present application do not make any limitations in this regard.

[0425] The transceiver 1013 can be used to support the reception or transmission of radio frequency signals between a communication device and a terminal. The transceiver 1013 can be connected to the antenna 1015. The transceiver 1013 includes a transmitter Tx and a receiver Rx. Specifically, one or more antennas 1015 can receive radio frequency signals. The receiver Rx of the transceiver 1013 is used to receive the radio frequency signals from the antenna, convert the radio frequency signals into digital baseband signals or digital intermediate frequency signals, and provide the digital baseband signals or digital intermediate frequency signals to the processor 1011 so that the processor 1011 can further process the digital baseband signals or digital intermediate frequency signals, such as demodulation processing and decoding processing. In addition, the transmitter Tx in the transceiver 1013 is also used to receive the modulated digital baseband signals or digital intermediate frequency signals from the processor 1011, convert the modulated digital baseband signals or digital intermediate frequency signals into radio frequency signals, and transmit the radio frequency signals through one or more antennas 1015. Specifically, the receiver Rx can selectively perform one-stage or multi-stage down-conversion processing and analog-to-digital conversion processing on the radio frequency signals to obtain digital baseband signals or digital intermediate frequency signals, and the sequence of the down-conversion processing and the analog-to-digital conversion processing can be adjusted. The transmitter Tx can selectively perform one-stage or multi-stage up-conversion processing and digital-to-analog conversion processing on the modulated digital baseband signals or digital intermediate frequency signals to obtain radio frequency signals, and the sequence of the up-conversion processing and the digital-to-analog conversion processing can be adjusted. The digital baseband signals and the digital intermediate frequency signals can be collectively referred to as digital signals.

[0426] The transceiver 1013 can also be referred to as a transceiver unit, a transceiver, a transceiver device, etc. Optionally, the devices used to implement the receiving function in the transceiver unit can be regarded as a receiving unit, and the devices used to implement the sending function in the transceiver unit can be regarded as a sending unit, that is, the transceiver unit includes a receiving unit and a sending unit. The receiving unit can also be referred to as a receiver, an input port, a receiving circuit, etc., and the sending unit can be referred to as a transmitter, a transmitter, or a transmitting circuit, etc.

[0427] It should be noted that Figure 10 The illustrated communication device 1000 can specifically be used to implement the steps implemented by the network device in the foregoing method embodiments and achieve the corresponding technical effects of the network device. Figure 10 For the specific implementation manners of the illustrated communication device 1000, reference can be made to the descriptions in the foregoing method embodiments, and details are not described herein one by one.

[0428] Please refer to Figure 11 , which is a schematic structural diagram of the communication device involved in the foregoing embodiments provided in the embodiments of the present application.

[0429] It can be understood that the communication device 110 includes, for example, modules, units, components, circuits, or interfaces, etc., which are appropriately configured together to execute the technical solutions provided in this application. The communication device 110 may be the terminal device or network device described above, or a component (such as a chip) in these devices, for implementing the methods described in the following method embodiments. The communication device 110 includes one or more processors 111. The processor 111 may be a general-purpose processor or a dedicated processor, etc. For example, it may 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 the data of software programs.

[0430] Optionally, in one design, the processor 111 may include a program 113 (sometimes also referred to as code or instructions), and the program 113 can be run on the processor 111, so that the communication device 110 executes the methods described in the following embodiments. In another possible design, the communication device 110 includes a circuit ( Figure 11 not shown).

[0431] Optionally, the communication device 110 may include one or more memories 112, on which there is a program 114 (sometimes also referred to as code or instructions), and the program 114 can be run on the processor 111, so that the communication device 110 executes the methods described in the above method embodiments.

[0432] Optionally, the processor 111 and / or the memory 112 may include AI modules 117, 118, and the AI modules are used to implement AI-related functions. The AI modules can be implemented in a software, hardware, or a combination of software and hardware manner. For example, the AI module may include a radio intelligence control (RIC) module. For example, the AI module may be a near-real-time RIC or a non-real-time RIC.

[0433] Optionally, data may also be stored in the processor 111 and / or the memory 112. The processor and the memory may be provided separately or integrated together.

[0434] Optionally, the communication device 110 may further include a transceiver 115 and / or an antenna 116. The processor 111 is sometimes also referred to as a processing unit, which controls the communication device (such as a RAN node or a terminal). The transceiver 115 is sometimes also referred to as a transceiver unit, a transceiver, a transceiver circuit, or a transceiver, etc., and is used to implement the transceiver function of the communication device through the antenna 116.

[0435] Among them, Figure 7 the processing unit 701 shown may be the processor 111. Figure 7 the transceiver unit 702 shown may be a communication interface, and this communication interface may be Figure 11 the transceiver 115 in, and this transceiver 115 may include an input interface and an output interface. Alternatively, this transceiver 115 may also be a transceiver circuit, and this transceiver circuit may include an input interface circuit and an output interface circuit.

[0436] The embodiments of the present application further provide 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 manners of the first communication device or the second communication device in the foregoing embodiments.

[0437] The embodiments of the present application further provide a computer program product (or referred to as a computer program). When the computer program product is executed by the processor, the processor executes the method of the possible implementation manners of the foregoing first communication device or second communication device.

[0438] The embodiments of the present application further provide a chip system, which includes at least one processor for supporting a communication device to implement the functions involved in the possible implementation manners of the foregoing communication device. Optionally, the chip system further includes an interface circuit, and the interface circuit provides program instructions and / or data for the at least one processor. In a possible design, the chip system may further include a memory for storing the necessary program instructions and data of the communication device. The chip system may be composed of chips or may include chips and other discrete devices, where the communication device may specifically be the first communication device or the second communication device in the foregoing method embodiments.

[0439] The embodiments of the present application further provide a communication system, and the network system architecture includes the first communication device and the second communication device in any of the foregoing embodiments.

[0440] In several embodiments provided by the present application, it should be understood that the disclosed systems, devices, and methods may be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division, and there may be other division methods in actual implementation. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Another point, the displayed or discussed couplings or direct couplings or communication connections to each other may be indirect couplings or communication connections through some interfaces, devices, or units, and may be in electrical, mechanical, or other forms.

[0441] The unit described as a separation component may or may not be physically separated. The component shown as a unit may or may not be a physical unit, that is, it may be located in one place or may be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0442] In addition, each functional unit in various embodiments of the present application may be integrated in a processing unit, may exist separately as individual physical units, or two or more units may be integrated in one unit. The above-mentioned integrated unit may 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 may be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present application, in essence, or the part that makes a contribution, or all or part of the technical solution, may be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present application. The foregoing storage medium includes: various media such as a USB flash drive, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk, or an optical disc that can store program codes.

Claims

1. A communication method, characterized in that, the method is applied to a first communication device, and the method includes: generating a first artificial intelligence (AI) model; sending first information to a second communication device, where the first information is used to determine the first AI model.

2. The method according to claim 1, characterized in that, the first AI model is a dedicated AI model.

3. The method according to claim 2, characterized in that, before generating the first AI model, it includes: receiving second information from the second communication device; generating the first AI model includes: generating the first AI model according to the second information.

4. The method according to claim 3, characterized in that, the second information includes at least one of the following: channel state information between the first communication device and the second communication device, the local computing power state of the second communication device.

5. The method according to claim 4, characterized in that, the second information may further include at least one of the following: location information of the second communication device, behavior information of the second communication device, local data information and label information of the second communication device.

6. The method according to any one of claims 1 to 5, characterized in that, the method further includes: receiving third information from a third communication device; sending the third information to the second communication device, where the third information is used to determine a third AI model, and the third AI model is a general AI model.

7. The method according to any one of claims 1 to 6, characterized in that, after sending the first information to the second communication device, the method further includes: receiving fourth information from the second communication device, where the fourth information is used to indicate one of the AI models in a list of one or more AI models.

8. The method according to claim 7, characterized in that, the list of one or more AI models includes a list of one or more dedicated AI models and / or a list of general AI models.

9. A communication method, characterized in that, the method is applied to a second communication device, and the method includes: receiving first information from a first communication device; determining a first AI model according to the first information.

10. The method according to claim 9, characterized in that, the first AI model is a dedicated AI model.

11. The method according to claim 10, characterized in that, before receiving the first information from the first communication device, the method further includes: sending second information to the first communication device, where the second information is used to generate the first AI model.

12. The method according to claim 11, characterized in that, the second information includes at least one of the following: channel state information between the first communication device and the second communication device, the local computing power state of the second communication device.

13. The method according to claim 12, characterized in that, the second information further includes at least one of the following: The location information of the second communication device, the behavior information of the second communication device, the local data information of the second communication device, and the label information.

14. The method according to any one of claims 10 to 13, wherein, the method further comprises: receiving third information from the first communication device, the third information being used to determine a third AI model, and the third AI model being a general AI model.

15. The method according to any one of claims 10 to 14, wherein, after determining the first AI model according to the first information, the method further comprises: sending fourth information to the first communication device, the fourth information being used to indicate one of the AI models in a list of one or more AI models.

16. The method according to claim 15, wherein, the list of one or more AI models includes a list of one or more dedicated AI models and / or a list of general AI models.

17. The method according to any one of claims 1 to 16, wherein, the first information is used to determine the category to which the second communication device belongs.

18. The method according to any one of claims 1 to 17, wherein, the first AI model is included in a first group of AI models, the first group of AI models further includes the second AI model, the first AI model is deployed on the first communication device, and the second AI model is deployed on the second communication device; wherein, the input of the second AI model includes the output of the first AI model, or the input of the first AI model includes the output of the second AI model.

19. The method according to claim 18, wherein, the first information is further used to determine the second AI model in the first group of AI models.

20. The method according to any one of claims 1 to 19, wherein, the first information includes an identifier of the first AI model, and the identifier is used to determine the first AI model in a list including one or more dedicated AI models.

21. The method according to claim 20, wherein, before sending the first information to the second communication device, the method further comprises: sending the list of one or more dedicated AI models to the second communication device.

22. The method according to any one of claims 1 to 19, wherein, the first information includes model parameters and / or model structure information of the first AI model.

23. The method according to claim 22, wherein, the first information further includes an identifier of the first AI model.

24. The method according to any one of claims 1 to 23, wherein, the first information is information sent periodically.

25. A communication device, wherein, comprises a module for performing the method according to any one of claims 1 to 18.

26. A communication device, wherein, Comprising at least one processor, the at least one processor being coupled to a memory; the at least one processor is configured to execute the method according to any one of claims 1 to 18.

27. The communication device according to claim 20, wherein, the communication device is a chip or a chip system.

28. A readable storage medium, wherein, the storage medium stores a computer program or instructions, and when the computer program or instructions are executed by a communication device, the method according to any one of claims 1 to 18 is implemented.

Images