Communication method and related apparatus

Abstract

A communication method and a related apparatus. In the method, upon determining a first sub-model among at least two sub-models comprised in a first model, a first communication apparatus can process first input data on the basis of the first sub-model to obtain first output data. Therefore, compared with a process in which a communication apparatus uses a complete model (i.e., the first model) to process input data to obtain output data, in the described solution, the first communication apparatus uses one of the sub-models comprised in the model to process input data to obtain output data, so that the data processing complexity can be reduced, and the data processing delay can also be reduced, thereby improving data processing efficiency. In some implementation processes, no other N-M model units need to participate in the process in which the first communication apparatus processes the first input data to obtain the first output data, so that the computing power consumption and power consumption of communication apparatuses can be reduced.

Description

Communication method and related apparatus

[0001] This application claims priority from the Chinese Patent Application No. 202410904349.1 filed on July 5, 2024, and entitled "A communication method and related apparatus", the content of which is incorporated herein by reference in its entirety. TECHNICAL FIELD

[0002] The present application relates to the field of communication, and in particular, to a communication method and related apparatus. BACKGROUND

[0003] With the development of communication technology, in a communication system, in addition to the traditional communication service, the service performed by the communication device can also include other new services, such as artificial intelligence (AI) services. Generally, a communication system capable of processing AI services can also be referred to as an AI system.

[0004] Currently, a communication device can be used as a participating node of an AI system, and the computing power of the communication device is applied to a certain link of the AI system. Generally, the AI function introduced in the communication network needs to rely on a model to be implemented. For example, the communication device can process input data through a model to obtain output data. Increasing the complexity of the model (such as increasing the number of parameters of the model, increasing the number of neural network layers included in the model, etc.) can effectively improve the performance of the output data.

[0005] However, in the case of gradually increasing the complexity of the model, it is possible to cause the above-mentioned data processing process to consume a relatively long time, thereby affecting the data processing efficiency. SUMMARY

[0006] The present application provides a communication method and related apparatus, which can reduce the complexity of data processing while also reducing the data processing delay, so as to improve the data processing efficiency.

[0007] The first aspect of the present application provides a communication method, which is performed by a first communication device. The first communication device can be a communication apparatus (e.g., a terminal device or a network device), or the first communication device can be a part of the communication apparatus (e.g., a circuit or a chip responsible for communication functions (e.g., a Modem chip (also referred to as a baseband chip), a system on chip (SoC) chip, such as an SoC chip including a modem core, or a system in package (SIP) chip), etc.), or the first communication device can also be a logic module or software capable of implementing all or part of the functions of the communication apparatus. In the method, the first communication device determines a first sub-model, the first sub-model including M model units, the M model units being part of N model units, M being a positive integer, and N being greater than M; wherein the N model units are used to determine a first model, the first model including at least two sub-models, the first sub-model being one of the at least two sub-models; and the first communication device processes first input data based on the first sub-model to obtain first output data.

[0008] Based on the above scheme, after determining the first sub-model of the at least two sub-models included in the first model, the first communication device can process the first input data based on the first sub-model to obtain the first output data. Thus, compared with the process of processing the input data based on the complete model (i.e., the first model) to obtain the output data, in the above scheme, the first communication device processes the input data based on one of the sub-models included in the model to obtain the output data, which can reduce the complexity of data processing and also reduce the data processing delay, thereby improving the data processing efficiency.

[0009] In addition, the first model is determined by the N model units, and the first sub-model is determined by the M model units of the N model units, and M is less than N. In this way, in the process of processing the first input data to obtain the first output data, the first communication device can not need the participation of the other N-M model units, which can reduce the computing power consumption and power consumption of the communication device.

[0010] In the present application, the model can include an AI model, a neural network model, an AI neural network model, a machine learning model, or an AI processing model. Similarly, the sub-model can include an AI sub-model, a neural network sub-model, an AI neural network sub-model, a machine learning sub-model, or an AI processing sub-model.

[0011] It should be understood that one model can be understood as an overall model, any sub-model included in the one model can be a partial model, and the any sub-model can include one or more model components. In other words, the models involved in the present application can be implemented in three granularities, including an overall model, a partial model, and a model component. In the above scheme, the overall model is denoted as a model (e.g., a first model), the partial model is denoted as a sub-model (e.g., at least two sub-models included in the first model, a first sub-model, etc.), and the model component is denoted as a model unit (e.g., N model units, M model units, etc.). Among them, these terms can be replaced by other descriptions.

[0012] As an example, the overall model can be a model, the partial model can be denoted as a sub-model group, and the model component can be denoted as a sub-model. For example, in the above scheme, the sub-model can be replaced by a sub-model group (e.g., at least two sub-models can be replaced by at least two sub-model groups, and the first sub-model can be replaced by a first sub-model group), and the model unit can be replaced by a sub-model (e.g., N model units can be replaced by N sub-models, and M model units can be replaced by M sub-models).

[0013] As another example, the overall model can be a joint model, the partial model can be denoted as a model, and the model component can be denoted as a sub-model. For example, in the above scheme, the model can be replaced by a joint model (e.g., the first model can be replaced by a first joint model), the sub-model can be replaced by a model (e.g., at least two sub-models can be replaced by at least two models, and the first sub-model can be replaced by a first model), and the model unit can be replaced by a sub-model (e.g., N model units can be replaced by N sub-models, and M model units can be replaced by M sub-models).

[0014] It should be noted that in the at least two sub-models included in the first model, the one or more model units and the connection order included in any sub-model are different from the one or more model units and the connection order included in the other sub-model.

[0015] For example, the one or more model units included in any sub-model are different from the one or more model units included in the other sub-model.

[0016] For another example, the one or more model units included in a certain sub-model are the same as the one or more model units included in another sub-model. Moreover, the connection order between the one or more model units included in the one sub-model is different from the connection order between the one or more model units included in the other sub-model.

[0017] In the application, in the process of obtaining output data by the communication device based on the model or the sub-model to process the input data, the data processing can include one or more of inference, prediction, derivation, identification, decision-making. For example, the input data can be inference data, and the output data can be inference output data. For example, in the above process, the first communication device can perform one or more of inference, prediction, derivation, identification, decision-making on the first input data based on the first sub-model to obtain the first output data.

[0018] In a possible implementation of the first aspect, the method further includes: the first communication device sending or receiving first information, the first information being used to indicate the first sub-model or model information of the first sub-model.

[0019] Based on the above scheme, the first communication device can send the first information, so that the receiver of the first information can determine the sub-model or the model information of the sub-model of the first communication device based on the first information, and further enable the receiver to perform model management (such as model inference, model scheduling, model updating, model switching, or function fallback, etc.) on the sub-model of the first communication device based on the first information.

[0020] Alternatively, the first communication device can receive the first information, so that the first communication device can determine the sub-model or the model information of the sub-model of the first communication device based on the first information, and further enable the first communication device to perform data processing on the model deployed by itself based on the indication of the other communication device.

[0021] In a possible implementation of the first aspect, the method further includes: the first communication device receiving second information, the second information being used to indicate the first input data; wherein the first input data is used to determine the first sub-model.

[0022] Based on the above scheme, the first communication device can receive the second information indicating the first input data, and can determine the first sub-model from the at least two sub-models based on the first input data. In other words, the first input data can be used to determine one or more of the at least two sub-models, and the first communication device can select the first sub-model from the one or more sub-models to process the first input data.

[0023] Optionally, the first communication device can randomly select the first sub-model from the one or more sub-models. Alternatively, the first communication device can select the first sub-model from the one or more sub-models based on its own state information and / or requirements, so that the process of the first communication device processing the first input data can meet the state indicated by the state information and / or the requirements.

[0024] Exemplarily, the state information can include a communication state and / or an AI state of the first communication device. For example, the communication state information can include one or more of a geographical position, a moving speed, a signal coverage strength, a signal to interference plus noise ratio (SINR) change, a channel response change, a data throughput change, and the like of the communication device. For another example, the AI state information can include one or more of a (current) model performance, a data distribution of model input data, a data distribution of model output data, and the like.

[0025] Optionally, the second information can include the first input data, or the second information can include data collection configuration information of the first input data.

[0026] Optionally, for the first communication device, the first input data can be collected or acquired by the first communication device itself, so that the input data of the model can be determined without the indication of other communication devices, and the overhead can be reduced.

[0027] In a possible implementation of the first aspect, the method further includes: receiving, by the first communication device, third information, the third information being used to indicate the first model; or obtaining, by the first communication device, the first model based on training data.

[0028] Based on the above scheme, the first communication device can obtain the first model in any of the above manners, so as to improve the flexibility of the scheme implementation.

[0029] In a possible implementation of the first aspect, the method further includes: receiving, by the first communication device, fourth information, the fourth information being used to indicate a second sub-model or model information of the second sub-model; and processing, by the first communication device, the first input data based on the first sub-model to obtain first output data, including: in a case where the state information of the first communication device does not satisfy the second sub-model, processing, by the first communication device, the first input data based on the first sub-model to obtain the first output data.

[0030] Based on the above scheme, the first communication device can receive the fourth information indicating the second sub-model or the model information of the second sub-model, and in a case where the state information of the first communication device does not satisfy the second sub-model, the first communication device can perform data processing based on other sub-models (for example, the first sub-model), so as to improve the data processing efficiency and reduce the latency.

[0031] Similarly, the state information can include a communication state and / or an AI state of the first communication device, which can be referred to the foregoing description.

[0032] In a possible implementation of the first aspect, the method further includes: the first communication device sending fifth information, the fifth information being used to indicate that the state information of the first communication device does not satisfy the second sub-model.

[0033] Based on the above scheme, the first communication device can send the fifth information, so that the receiver of the fifth information can explicitly know that the state information of the first communication device does not satisfy the second sub-model, and explicitly know that the first communication device performs data processing based on other sub-models (for example, the first sub-model).

[0034] Optionally, in the case where it is determined that the state information of the first communication device does not satisfy the second sub-model, the first communication device can randomly select or select one of other sub-models based on the state information as a sub-model for data processing, other than the second sub-model. Alternatively, after the first communication device sends the fifth information, the receiver of the fifth information can send the first information (which can refer to the foregoing implementation) to the first communication device based on the fifth information, so that the first communication device explicitly selects the first sub-model as a sub-model for data processing based on the first information.

[0035] In a possible implementation of the first aspect, the method further includes: the first communication device sending sixth information, the sixth information being used to indicate the first output data.

[0036] Based on the above scheme, the first communication device can further send the sixth information, so that the receiver of the sixth information can obtain the first output data obtained by the first communication device through data processing, to realize feedback / indication of model output data.

[0037] The second aspect of the present application provides a communication method, which is performed by a second communication device. The second communication device can be a communication device (such as a terminal device or a network device), or the second communication device can be a part of the communication device (for example, a circuit or a chip responsible for communication functions (such as a Modem chip (also known as a baseband chip), a SoC chip, such as a SoC chip containing a modem core, or a SIP chip, etc.), or the second communication device can also be a logic module or software capable of realizing all or part of the functions of the communication device. In the method, the second communication device determines first information, the first information being used to indicate a first sub-model or model information of the first sub-model; wherein the first sub-model includes M model units, the M model units being part of N model units, M being a positive integer, and N being greater than M; the N model units are used to determine a first model, the first model including at least two sub-models, and the first sub-model being one of the at least two sub-models; and the second communication device sends the first information.

[0038] Based on the above scheme, the first information sent by the second communication device to the first communication device is used to indicate the first sub-model or the model information of the first sub-model, and then the first communication device can process the first input data based on the first sub-model to obtain the first output data. Thus, compared with the process of processing the input data by the communication device based on the complete model (i.e., the first model) to obtain the output data, in the above scheme, the first communication device processes the input data based on one of the sub-models included in the model to obtain the output data, which can reduce the complexity of data processing and also reduce the data processing delay, thereby improving the data processing efficiency.

[0039] In addition, the first model is determined by N model units, the first sub-model is determined by M model units of the N model units, and M is less than N. In this way, in the process of processing the first input data to obtain the first output data, the first communication device can not need the participation of other N-M model units, which can reduce the computing power consumption and power consumption of the communication device.

[0040] The third aspect of the present application provides a communication method, which is executed by a second communication device. The second communication device can be a communication device (such as a terminal device or a network device), or the second communication device can be a part of the communication device (for example, a circuit or a chip responsible for communication functions (such as a Modem chip (also known as a baseband chip), a SoC chip, such as a SoC chip containing a modem core, or a SIP chip, etc.), or the second communication device can also be a logic module or software that can realize all or part of the functions of the communication device. In the method, the second communication device receives first information, the first information is used to indicate a first sub-model or model information of the first sub-model; wherein the first sub-model includes M model units, the M model units are part of the N model units, M is a positive integer, and N is greater than M; the N model units are used to determine a first model, the first model includes at least two sub-models, and the first sub-model is one of the at least two sub-models; and the second communication device performs model management on the first sub-model based on the first information.

[0041] Based on the above scheme, the second communication apparatus receives first information from the first communication apparatus for indicating a first sub-model or model information of the first sub-model, where the first communication apparatus can process first input data based on the first sub-model to obtain first output data. Subsequently, the second communication apparatus performs model management (such as model inference, model scheduling, model updating, model switching, or function fallback, etc.) on the first sub-model based on the first information. Thus, compared with the process of the communication apparatus processing input data based on a complete model (i.e., the first model) to obtain output data, in the above scheme, the first communication apparatus processes input data based on one of the sub-models contained in the model to obtain output data, which can reduce the complexity of data processing and also reduce the data processing delay, thereby improving the data processing efficiency.

[0042] In addition, the first model is determined by N model units, the first sub-model is determined by M model units of the N model units, and M is less than N. In this way, in the process of processing the first input data to obtain the first output data, the first communication apparatus can not need to involve other N-M model units, which can reduce the computing power consumption and power consumption of the communication apparatus.

[0043] In a possible implementation of the second aspect or the third aspect, the method further includes: the second communication apparatus sending second information, the second information being used for indicating the first input data; and the first input data being used for determining the first sub-model.

[0044] Based on the above scheme, the second communication apparatus can send second information indicating the first input data, so that the first communication apparatus can determine the first sub-model from the at least two sub-models based on the first input data. In other words, the first input data can be used to determine one or more sub-models of the at least two sub-models, and the first communication apparatus can select the first sub-model for processing the first input data from the one or more sub-models.

[0045] In a possible implementation of the second aspect or the third aspect, the method further includes: the second communication apparatus sending third information, the third information being used for indicating the first model; or the first model being obtained by the first communication apparatus based on training data.

[0046] Based on the above scheme, the first communication apparatus can obtain the first model in any of the above ways, so as to improve the flexibility of the scheme implementation.

[0047] In a possible implementation of the second aspect or the third aspect, the method further includes: sending, by the second communication apparatus, fourth information used to indicate the second sub-model or model information of the second sub-model; and receiving, by the first communication apparatus, fifth information used to indicate that the state information of the first communication apparatus does not satisfy the second sub-model.

[0048] Based on the above scheme, after the second communication apparatus sends the fourth information used to indicate the second sub-model or model information of the second sub-model to the first communication apparatus, the second communication apparatus can receive the fifth information, so that the second communication apparatus can explicitly determine that the state information of the first communication apparatus does not satisfy the second sub-model, and explicitly determine that the first communication apparatus performs data processing based on other sub-models (for example, the first sub-model). For example, in the case that the first communication apparatus determines that the state information of the first communication apparatus does not satisfy the second sub-model, the first communication apparatus can perform data processing based on other sub-models (for example, the first sub-model) to improve data processing efficiency and reduce latency.

[0049] In a possible implementation of the second aspect or the third aspect, the method further includes: receiving, by the second communication apparatus, sixth information used to indicate the first output data.

[0050] Based on the above scheme, the second communication apparatus can further receive the sixth information, so that the second communication apparatus can obtain the first output data obtained by the first communication apparatus through data processing, to realize feedback / indication of model output data, and perform model management on the sub-models of the first communication apparatus through data processing based on the first output data.

[0051] In a possible implementation of the first aspect or the second aspect or the third aspect, the model information of any sub-model (or at least one sub-model) is preconfigured.

[0052] Based on the above scheme, in the at least two sub-models included in the first model, the model information of any sub-model (or at least one sub-model) can be preconfigured, which can reduce implementation complexity and communication overhead.

[0053] Optionally, the model information of any sub-model (or at least one sub-model) can be configured by a network device, to improve flexibility of implementation of the scheme.

[0054] Optionally, in the at least two sub-models included in the first model, the model information of any sub-model includes at least one of: an index corresponding to the any sub-model, a use case corresponding to the any sub-model, an AI feature corresponding to the any sub-model, a function corresponding to the any sub-model, a task corresponding to the any sub-model, a data type of input data of the any sub-model, a data type of output data of the any sub-model, an identifier of one or more model units included in the any sub-model, a connection relationship between the one or more model units included in the any sub-model, or a target performance corresponding to the any sub-model.

[0055] In a possible implementation of the first aspect or the second aspect or the third aspect, in the at least two sub-models, the target performances corresponding to different sub-models are different.

[0056] Based on the above scheme, in the at least two sub-models included in the first model, the target performances corresponding to different sub-models are different, so that the first communication device (or the provider of the first information) can perform data processing on the corresponding sub-model based on the performance requirement or the performance indication to meet the performance requirement or the performance indication.

[0057] The fourth aspect of the present application provides a communication device, which is a first communication device, and the device includes a processing unit; the processing unit is configured to determine a first sub-model, the first sub-model includes M model units, the M model units are part of N model units, M is a positive integer, and N is greater than M; wherein the N model units are configured to determine a first model, the first model includes at least two sub-models, and the first sub-model is one of the at least two sub-models; the processing unit is further configured to process first input data based on the first sub-model to obtain first output data.

[0058] In the fourth aspect of the present application, the component modules of the communication device can also be configured to perform the steps performed in the various possible implementation manners of the first aspect and achieve the corresponding technical effects, which can be referred to the first aspect for details and will not be described here.

[0059] The fifth aspect of the present application provides a communication device, which is a second communication device, and the device includes a transceiver unit and a processing unit; the processing unit is configured to determine first information, the first information is used to indicate a first sub-model or model information of the first sub-model; wherein the first sub-model includes M model units, the M model units are part of N model units, M is a positive integer, and N is greater than M; the N model units are configured to determine a first model, the first model includes at least two sub-models, and the first sub-model is one of the at least two sub-models; and the transceiver unit is configured to send the first information.

[0060] In the fifth aspect of the present application, the constituent modules of the communication device can also be used to perform the steps performed in the various possible implementation manners of the second aspect and achieve the corresponding technical effects, which can be specifically referred to the second aspect and will not be described here again.

[0061] The sixth aspect of the present application provides a communication device, which is a second communication device. The communication device comprises a transceiver unit and a processing unit. The transceiver unit is configured to receive first information. The first information is used to indicate a first sub-model or model information of the first sub-model. The first sub-model comprises M model units, and the M model units are part of N model units. M is a positive integer, and N is greater than M. The N model units are used to determine a first model. The first model comprises at least two sub-models, and the first sub-model is one of the at least two sub-models. The processing unit is configured to perform model management on the first sub-model based on the first information.

[0062] In the sixth aspect of the present application, the constituent modules of the communication device can also be used to perform the steps performed in the various possible implementation manners of the third aspect and achieve the corresponding technical effects, which can be specifically referred to the third aspect and will not be described here again.

[0063] The seventh aspect of the present application provides a communication device comprising at least one processor configured to execute computer programs or instructions to enable the communication device to implement the method described in any one of the possible implementation manners of any one of the first aspect to the third aspect. Optionally, the communication device can comprise a memory (or the communication device can be externally connected to a memory), and the memory is configured to store the computer programs or instructions.

[0064] The eighth aspect of the present application provides a communication device comprising at least one logic circuit and an input and output interface. The logic circuit is configured to execute the method described in any one of the possible implementation manners of any one of the first aspect to the third aspect.

[0065] The ninth aspect of the present application provides a communication system comprising the first communication device and the second communication device.

[0066] The tenth aspect of the present application provides a computer readable storage medium configured to store one or more computer execution instructions. When the computer execution instructions are executed by a processor, the processor executes the method described in any one of the possible implementation manners of any one of the first aspect to the third aspect.

[0067] 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 in any possible implementation manner of any one of the first aspect to the third aspect.

[0068] The twelfth aspect of the present application provides a chip system, which comprises at least one processor for supporting the communication device to implement the method in any possible implementation manner of any one of the first aspect to the third aspect.

[0069] In a possible design, the chip system can further comprise a memory for storing necessary program instructions and data of the communication device. The chip system can be composed of a chip, or can comprise the chip and other discrete devices. Optionally, the chip system further comprises an interface circuit, which provides program instructions and / or data for the at least one processor.

[0070] The technical effects brought by any one of the fourth aspect to the twelfth aspect can be referred to the technical effects brought by different design manners of the first aspect to the third aspect, which will not be repeated here. BRIEF DESCRIPTION OF DRAWINGS

[0071] FIGS. 1a to 1c are schematic diagrams of a communication system provided by the present application;

[0072] FIGS. 2a to 2g are schematic diagrams of an AI processing process related to the present application;

[0073] FIG. 3 is an interaction schematic diagram of a communication method provided by the present application;

[0074] FIGS. 4a to 4d are some schematic diagrams of a model provided by the present application;

[0075] FIGS. 5 to 9 are schematic diagrams of a communication device provided by the present application. DETAILED DESCRIPTION

[0076] First, some terms in the embodiments of the present application are explained to facilitate understanding by those skilled in the art.

[0077] (1) Terminal device: can be a wireless terminal device capable of receiving network device scheduling and indication information, the wireless terminal device can be a device providing voice and / or data connectivity to a user, or a handheld device with wireless connection function, or other processing devices connected to a wireless modem.

[0078] A terminal device can communicate with one or more core networks or the Internet via a radio access network (RAN), and the terminal device can be a mobile terminal device, such as a mobile phone (or called "cellular" phone, mobile phone), a computer, and a data card, for example, which can be a portable, pocket, hand-held, computer- built-in, or vehicle-mounted mobile device that exchanges voice and / or data with a radio access network. For example, a personal communication service (PCS) phone, a cordless phone, a session initiation protocol (SIP) phone, a wireless local loop (WLL) station, a personal digital assistant (PDA), a tablet, a computer with wireless transceiver function, and the like. The wireless terminal device can also be referred to as a system, a subscriber unit, a subscriber station, a mobile station (MS), a remote station, an access point (AP), a remote terminal, an access terminal, a user terminal, a user agent, a subscriber station (SS), customer premises equipment (CPE), a terminal, user equipment (UE), a mobile terminal (MT), and the like.

[0079] By way of example and not limitation, in embodiments of the present application, the terminal device can also be a wearable device. The wearable device can also be referred to as a smart wearable device or a smart wearable device, etc., which is a general term for devices that can be designed and developed by applying wearable technology to daily wear, such as glasses, gloves, watches, clothing, and shoes, etc. The wearable device is a portable device that can be directly worn on the body or integrated into the user's clothes or accessories. The wearable device is not only a hardware device, but also a powerful function realized through software support and data interaction, cloud interaction. The general wearable smart device includes a full function, large size, and can realize complete or partial functions without relying on a smart phone, such as smart watches or smart glasses, etc., and focuses on a certain application function, and needs to cooperate with other devices such as a smart phone, such as various smart wristbands, smart helmets, smart jewelry, etc. for monitoring vital signs.

[0080] The terminal can also be a drone, a robot, a terminal in device-to-device (D2D) communication, a terminal in vehicle to everything (V2X), a virtual reality (VR) terminal device, an augmented reality (AR) terminal device, a wireless terminal in industrial control, a wireless terminal in self driving, a wireless terminal in telemedicine or telehealth services, 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.

[0081] In addition, the terminal device can also be a terminal device in a communication system evolved after the 5th generation (5G) communication system (such as 5G Advanced or 6th generation (6G) communication system, etc.) or a terminal device in a future evolved public land mobile network (PLMN), etc. For example, 5G Advanced or 6G network can further expand the form and function of 5G communication terminal, and 6G terminal includes but is not limited to vehicle, cellular network terminal (integrating satellite terminal function), drone, internet of things (IoT) device.

[0082] In the embodiments of the present application, the terminal device can also obtain an artificial intelligence (AI) service provided by the network device. Optionally, the terminal device can also have AI processing capability.

[0083] (2) Network device: can be a device in a wireless network, for example, the network device can be a RAN node (or device) for accessing the terminal device to the wireless network, which can also be referred to as a base station. At present, some examples of RAN devices are: base station, evolved NodeB (eNodeB), base station gNB (gNodeB) in 5G communication system, transmission reception point (transmission reception point or transmit / receive point, TRP), evolved Node B (eNB), radio network controller (radio network controller, RNC), Node B (Node B, NB), home base station (for example, home evolved Node B, or home Node B, HNB), baseband unit (baseband unit, BBU) or wireless fidelity (wireless fidelity, Wi-Fi) access point (AP) and the like. In addition, in one network structure, the network device can include a central unit (central unit, CU) node, or a distributed unit (distributed unit, DU) node, or a RAN device including a CU node and a DU node.

[0084] 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 wireless controller in a cloud radio access network (cloud radio access network, CRAN) scenario. The RAN node can also be a server, a wearable device, a vehicle or a vehicle-mounted device, etc. For example, the access network device in vehicle external connection (V2X) technology can be a road side unit (road side unit, RSU).

[0085] In another possible scenario, multiple RAN nodes cooperate to assist a terminal to implement wireless access, and different RAN nodes respectively implement part of functions of a base station. For example, a RAN node can be a CU, a DU, a CU-control plane (CP), a CU-user plane (UP), or a radio unit (RU), etc. The CU and the DU can be separately configured, or can be included in the same network element, for example, in a baseband unit (BBU). The RU can be included in a radio frequency device or a radio frequency unit, for example, in a remote radio unit (RRU), an active antenna unit (AAU), a radio head (RH), or a remote radio head (RRH).

[0086] In different systems, the CU (or CU-CP and CU-UP), the DU, or the RU can 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 referred to as an O-CU (open CU), the DU can also be referred to as an O-DU, the CU-CP can also be referred to as an O-CU-CP, the CU-UP can also be referred to as an O-CU-UP, and the RU can also be referred to as an O-RU. For the convenience of description, the CU, the CU-CP, the CU-UP, the DU, and the RU are taken as examples for description in this application. Any one of the CU (or the CU-CP, the CU-UP), the DU, and the 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.

[0087] The communication between the access network device and the terminal device complies with a certain protocol layer structure. The protocol layer can include a control plane protocol layer and a user plane protocol layer. The control plane protocol layer can include at least one of a radio resource control (RRC) layer, a packet data convergence protocol (PDCP) layer, a radio link control (RLC) layer, a media access control (MAC) layer, or a physical (PHY) layer, etc. The user plane protocol layer can include at least one of a service data adaptation protocol (SDAP) layer, a PDCP layer, an RLC layer, a MAC layer, or a physical layer, etc.

[0088] For the correspondence between the network elements in the ORAN system and the protocol layer functions that can be implemented by the network elements, refer to Table 1 below.

[0089] Table 1

[0090] The network device can be another device that provides a wireless communication function for a terminal device. Embodiments of the present application do not limit the specific technology and specific device form adopted by the network device. For the convenience of description, embodiments of the present application do not limit.

[0091] The network device can also include a core network device, for example, a mobility management entity (MME) in a fourth generation (4G) network, a home subscriber server (HSS), a serving gateway (S-GW), a policy and charging rules function (PCRF), a public data network gateway (PDN gateway or P-GW), a network element such as an access and mobility management function (AMF) in a 5G network, a user plane function (UPF), or a session management function (SMF). In addition, the core network device can also include other core network devices in the 5G network and the next generation network of the 5G network.

[0092] In embodiments of the present application, the network device described above can also be an AI-capable network node that can provide AI services for terminals or other network devices, for example, AI nodes, computing power nodes, AI-capable RAN nodes, AI-capable core network elements, etc. on the network side (access network or core network).

[0093] In embodiments of the present application, the device for implementing the function of the network device can be a network device or a device capable of supporting the network device to implement the function, such as a chip system, which can be arranged in the network device. In the technical solutions provided in embodiments of the present application, the device for implementing the function of the network device is taken as an example to describe the technical solutions provided in embodiments of the present application.

[0094] (3) Configuration and pre-configuration: in this application, configuration and pre-configuration will be used at the same time. Among them, configuration refers to that the network device / server sends some parameter configuration information or parameter values to the terminal through messages or signaling, so that the terminal determines the communication parameters or resource in transmission according to the values or information. Pre-configuration is similar to configuration, which can be parameter information or parameter values agreed by the network device / server and the terminal device in advance, or parameter information or parameter values adopted by the base station / network device or the terminal device according to the standard protocol, or parameter information or parameter values pre-stored in the base station / server or the terminal device. This application does not limit it.

[0095] Further, these values and parameters can be changed or updated.

[0096] (4) The terms "system" and "network" in the embodiments of the present application can be used interchangeably. "Multiple" means two or more. "And / or" describes the association relationship of the associated objects, which means that there can be three relationships, for example, A and / or B, which means that A exists alone, A and B exist at the same time, and B exists alone, where A and B can be singular or plural. The character " / " generally represents an "or" relationship between the associated objects. "At least one of the following" or similar expressions means any combination of these items, including any combination of single item or multiple items. For example, "at least one of A, B and C" includes A, B, C, AB, AC, BC or ABC. In addition, unless otherwise specified, the ordinal numbers "first", "second", etc. 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 of multiple objects.

[0097] (5) In the embodiments of the present application, "sending" and "receiving" represent the direction of signal transmission. For example, "sending information to XX" can be understood as that the destination of the information is XX, which can include direct sending through the air interface, or indirect sending through the air interface by other units or modules. "Receiving information from YY" can be understood as that the source of the information is YY, which can include direct receiving from YY through the air interface, or indirect receiving from YY through the air interface by other units or modules. "Sending" can also be understood as "output" of chip interface, and "receiving" can also be understood as "input" of chip interface.

[0098] In other words, sending and receiving can be carried out between devices, such as between network devices and terminal devices, or within devices, such as between components, modules, chips, software modules or hardware modules within devices through buses, wires or interfaces.

[0099] It can be understood that the information can be processed, such as encoding and modulation, between the source end and the destination end of the information transmission, but the destination end can understand the effective information from the source end. Similar expressions in this application can be similarly understood, and will not be repeated here.

[0100] (6) In the embodiments of the present application, “indication” can include direct indication and indirect indication, and can also include explicit indication and implicit indication. The information indicated by certain information (indication information described below) is referred to as to-be-indicated information. In the implementation process, there are many ways to indicate the to-be-indicated information, for example, but not limited to, the to-be-indicated information can be directly indicated, such as the to-be-indicated information itself or the index of the to-be-indicated information. The to-be-indicated information can also be indirectly indicated by indicating other information, where the other information and the to-be-indicated information have an association relationship. The to-be-indicated information can also be indicated only by a part, and the other part of the to-be-indicated information is known or agreed in advance. For example, the indication of a specific information can be achieved by means of the arrangement order of each information agreed in advance (for example, protocol predefined), 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 to-be-indicated information, and for the receiver of the indication information, the indication information can be used to determine the to-be-indicated information.

[0101] In the present application, the same or similar parts of each embodiment can be mutually referenced unless otherwise specified. In the present application, the terms and / or descriptions of different embodiments and the methods / designs / implementation manners in each embodiment are consistent and can be mutually referenced if not specially specified and there is no logical conflict. The technical features of different embodiments and the methods / designs / implementation manners in each embodiment can be combined to form new embodiments, methods, or implementation manners according to their inherent logical relationship. The implementation manners of the present application described below do not constitute a limitation on the protection scope of the present application.

[0102] 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 6G, etc.). The communication system includes at least one network device and / or at least one terminal device.

[0103] Please refer to FIG. 1a, which is a schematic diagram of a communication system in the present application. In FIG. 1a, a network device and six terminal devices are exemplarily shown, which are terminal device 1, terminal device 2, terminal device 3, terminal device 4, terminal device 5 and terminal device 6. In the example shown in FIG. 1a, the terminal device 1 is exemplarily taken as a smart tea cup, the terminal device 2 is exemplarily taken as a smart air conditioner, the terminal device 3 is exemplarily taken as a smart gas station, the terminal device 4 is exemplarily taken as a vehicle, the terminal device 5 is exemplarily taken as a mobile phone, and the terminal device 6 is exemplarily taken as a printer.

[0104] As shown in FIG. 1a, the sending entity of the AI configuration information can be the network device. The receiving entity of the AI configuration information can be the terminal devices 1-6. In this case, the network device and the terminal devices 1-6 form a communication system, in which the terminal devices 1-6 can send data to the network device, and the network device receives the data sent by the terminal devices 1-6. The network device can send configuration information to the terminal devices 1-6.

[0105] Exemplarily, in FIG. 1a, the terminal devices 4-6 can also form a communication system. Among them, the terminal device 5 acts as a network device, i.e., the sending entity of the AI configuration information; the terminal devices 4 and 6 act as terminal devices, i.e., the receiving entity of the AI configuration information. For example, in a vehicle-to-everything system, the terminal device 5 sends AI configuration information to the terminal devices 4 and 6, and receives data sent by the terminal devices 4 and 6; correspondingly, the terminal devices 4 and 6 receive the AI configuration information sent by the terminal device 5, and send data to the terminal device 5.

[0106] Taking the communication system shown in FIG. 1a as an example, in addition to performing communication-related services, different devices (including network devices, network devices and terminal devices, and / or terminal devices and terminal devices) can also perform AI-related services.

[0107] As shown in FIG. 1b, taking a network device as a base station as an example, the base station can perform communication-related services and AI-related services with one or more terminal devices, and different terminal devices can also perform communication-related services and AI-related services.

[0108] As shown in FIG. 1c, taking a terminal device including a television and a mobile phone as an example, the television and the mobile phone can also perform communication-related services and AI-related services.

[0109] The technical solutions provided in the present application can be applied to a wireless communication system (for example, the system shown in FIG. 1a, FIG. 1b or FIG. 1c), for example, an AI network element can be introduced in the communication system provided in the present application to implement part or all of the AI related operations. The AI network element can also be referred to as an AI node, an AI device, an AI entity, an AI module, an AI model, or an AI unit, etc. The AI network element can be built-in in a network element of the communication system. For example, the AI network element can be an AI module built-in in an access network device, a core network device, a cloud server, or an operation, administration and maintenance (OAM) to implement AI related functions. The OAM can be a network management of the core network device and / or a network management of the access network device. Alternatively, the AI network element can also be a network element independently arranged in the communication system. Optionally, an AI entity can also be included in a terminal or a chip built-in in the terminal to implement AI related functions.

[0110] Optionally, in the communication system, the AI application cases can include but are not limited to: channel state information (CSI) feedback enhancement, beam management enhancement, positioning enhancement, network energy saving, load balancing, and mobility optimization. The following will be described respectively.

[0111] 1. CSI feedback enhancement

[0112] CSI is the channel property of the communication link, and is the channel quality information reported by the terminal device to the network device. The terminal device reports the channel quality information to the network device, so as to select a suitable modulation and coding scheme (MCS) for the terminal device, so that the wireless channel can be adapted to the change. For example, the terminal device performs channel estimation according to the received channel state information-reference signal (CSI-RS), and then feeds back the channel quality information to the network device. The information is used as the input of the model of the network device, so that the network device can implement AI model training. By applying AI to CSI feedback enhancement, the overhead can be reduced, the accuracy can be improved, and prediction can be realized.

[0113] CSI-RS feedback enhancement can include at least one sub-function, such as: CSI compression, CSI prediction, and CSI-RS configuration signaling reduction, respectively. CSI compression can further include CSI compression in at least one of spatial, time, and frequency domains.

[0114] 2. Beam management enhancement

[0115] Beam management enhancement is mainly to find the strongest transmit / receive beam pair. AI-based sparse beam prediction can improve accuracy. According to AI training and inference, it can include AI sparse beam prediction on the network side and AI sparse beam prediction on the terminal device side. Taking AI sparse beam prediction on the terminal device side as an example, the pre-trained AI model on the terminal device side can be delivered by the network side or pre-stored on the terminal device side. In the training phase, the network device scans all possible beams, and then the network reports the transmit beam pattern to the terminal device. When the model training is completed, the network device only needs to scan a small part of the beam, and then the terminal device feeds back the inference result to the network device. AI-based beam management can realize, for example, beam prediction in time and / or spatial domain to reduce overhead and delay and improve beam selection accuracy.

[0116] Beam management enhancement can include at least one sub-function, such as: beam scanning matrix prediction, and optimal beam prediction, respectively.

[0117] 3. Positioning enhancement (such as positioning accuracy enhancement)

[0118] In line of sight (LOS) or non-line of sight (NLOS) scenarios, AI-based positioning can improve positioning accuracy with a smaller number of TRP antennas. Positioning enhancement can include at least one sub-function, such as: access network device-based positioning enhancement, positioning management function network element-based positioning enhancement, and terminal device-based positioning enhancement, respectively.

[0119] 4. Network energy saving

[0120] Network energy saving can be achieved by cell activation / deactivation, load reduction, coverage improvement, or other RAN setting adjustment. AI technology can be used to optimize energy saving decisions by utilizing data collected in the RAN network. AI algorithms can predict the energy efficiency and load status of the next period, which can be used to assist in decision-making for cell activation / deactivation to save energy. Based on the predicted load, the system can dynamically configure the energy saving strategy to maintain the balance between system performance and energy efficiency, and reduce energy consumption.

[0121] 5. Load balancing

[0122] Load balancing can make the load evenly distributed among cells and between areas within a cell, or divert part of the traffic from congested cells, or split users among cells, carriers, or access modes to improve network performance. AI models can be used to improve load balancing performance, such as inputting various measurements and feedbacks of terminal devices and network nodes, historical data, etc. into AI models to improve load balancing performance, which can provide a higher quality of user experience and improve system capacity.

[0123] 6. Mobility management

[0124] Mobility management is a solution to ensure service continuity during terminal device movement by minimizing dropped calls, radio link failures (RLF), unnecessary handovers, and ping-pong effects. AI can enhance mobility management, such as reducing the probability of unexpected events, predicting terminal device location / mobility / performance, and traffic steering, etc.

[0125] It should be understood that the definitions of the above technical terms are only examples. For example, as technology continues to evolve, the scope of the above definitions can also change, and the embodiments of the present application are not limited.

[0126] For example, an AI function can include multiple AI sub-functions.

[0127] Optionally, AI application cases are also referred to as AI application scenarios or AI functions.

[0128] As described above, AI can be widely used in CSI feedback enhancement, beam management, positioning enhancement, energy saving, mobility enhancement, and load balancing to improve network performance. AI models can be deployed on the network side and / or the terminal device side, and the training of AI models relies on the collection of training data, which can come from terminal device measurements and feedbacks.

[0129] The following will briefly introduce the concepts that may be involved in the present application.

[0130] AI can enable machines to have human intelligence, for example, enabling machines to apply computer hardware and software to simulate certain intelligent behaviors of humans. To achieve artificial intelligence, a machine learning method can be employed. In the machine learning method, a model is learned (or trained) by a machine using training data. The model represents a mapping between an input and an output. The learned model can be used for inference (or prediction), i.e., the model can be used to predict an output corresponding to a given input. The output can also be referred to as an inference result (or a prediction result).

[0131] Machine learning can include supervised learning, unsupervised learning, and reinforcement learning. The unsupervised learning can also be referred to as non-supervised learning.

[0132] Supervised learning learns a mapping relationship from sample values to sample labels using a machine learning algorithm according to the sample values and the sample labels that have been collected, and uses an AI model to express the learned mapping relationship. The process of training a machine learning model is a process of learning such a mapping relationship. In the training process, a sample value is input into the model to obtain a predicted value of the model, and the model parameters are optimized by calculating the error between the predicted value of the model and the sample label (ideal value). After the mapping relationship is learned, the learned mapping can be used to predict new sample labels. The learned mapping relationship of supervised learning can include linear mapping or non-linear mapping. According to the type of label, the learned task can be divided into classification tasks and regression tasks.

[0133] Unsupervised learning uses algorithms to discover the internal patterns of samples according to the collected sample values. In unsupervised learning, a class of algorithms uses the sample itself as a supervision signal, i.e., the model learns a mapping relationship from the sample to the sample, which is called self-supervised learning. In the training process, the model parameters are optimized by calculating the error between the predicted value of the model and the sample itself. Self-supervised learning can be used for signal compression and decompression recovery applications, and common algorithms include autoencoders and generative adversarial networks.

[0134] Reinforcement learning is different from supervised learning, and is a class of algorithms that learn a strategy to solve a problem by interacting with the environment. Unlike supervised and unsupervised learning, the reinforcement learning problem does not have clear "correct" action label data, and the algorithm needs to interact with the environment to obtain reward signals from the environment feedback, and then adjust the decision action to obtain a larger reward signal value. In the following power control, the reinforcement learning model adjusts the downlink transmission power of each user according to the system total 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 a mapping relationship between the environment state and the optimal (e.g., optimal) decision action. However, because the "correct action" label cannot be obtained in advance, the network cannot be optimized by calculating the error between the action and the "correct action". Reinforcement learning training is achieved through iterative interaction with the environment.

[0135] Neural networks (NNs) are a specific model in machine learning techniques. According to the general approximation theorem, neural networks can theoretically approximate any continuous function, thus enabling them to learn arbitrary mappings. Traditional communication systems rely on extensive expert knowledge to design communication modules, while deep learning communication systems based on neural networks can automatically discover hidden pattern structures from large datasets, establish mapping relationships between data, and achieve performance superior to traditional modeling methods.

[0136] The idea behind neural networks comes from the neuronal structure of the brain. For example, each neuron performs a weighted summation of its input values ​​and outputs the result through an activation function.

[0137] Figure 2a shows a schematic diagram of a neuron structure. Assume the input to the neuron is x = [x0, x1, ..., x...]. n The weights corresponding to each input are w = [w0, w1, ..., w] n ], where n is a positive integer, w i and x i It can be any possible type, such as a decimal, an integer (e.g., 0, a positive integer, or a negative integer), or a complex number. i As x i The weights are used to assign weights to x. i Weighting is applied. The bias for the weighted sum of the input values ​​is, for example, b. Activation functions can take many forms. Suppose the activation function of a neuron is: y = f(z) = max(0, z), then the output of that neuron is: For example, if the activation function of a neuron is y = f(z) = z, then the output of that neuron is: Here, b can be any possible type, such as a decimal, an integer (e.g., 0, a positive integer, or a negative integer), or a complex number. The activation functions of different neurons in a neural network can be the same or different.

[0138] 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 expressiveness of the neural network can be improved, providing a more powerful information extraction and abstract modeling capability for complex systems. 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 referred to as the width of the layer. In an implementation, the 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 transmits the processing result to the output layer to obtain the output result of the neural network. In another implementation, the 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 transmits the processing result to the intermediate hidden layer. The hidden layer calculates the received processing result to obtain a calculation result, and transmits the calculation result to the output layer or the next adjacent hidden layer. Finally, the output layer obtains the output result of the neural network. The neural network can include one hidden layer, or include multiple sequentially connected hidden layers, without limitation.

[0139] The neural network is, for example, a deep neural network (DNN). According to the construction manner of the network, the DNN can include a feedforward neural network (FNN), a convolutional neural network (CNN), and a recurrent neural network (RNN).

[0140] FIG. 2b is a schematic diagram of an FNN network. The FNN network is characterized in that the neurons in adjacent layers are completely connected to each other. This characteristic makes the FNN usually require a large amount of storage space and cause high computational complexity.

[0141] The CNN is a neural network specially designed to process data with a similar grid structure. For example, time series data (e.g., discrete sampling on a time axis) and image data (e.g., two-dimensional discrete sampling) can be considered as data with a similar grid structure. Instead of using all input information at once for operation, the CNN uses a fixed-size window to extract part of the information for convolution operation, which greatly reduces the calculation amount of model parameters. In addition, according to the different types of information extracted by the window (such as people and objects in the same image), each window can use different convolution kernel operations, which enables the CNN to better extract the features of the input data.

[0142] RNN is a kind of neural network using feedback time series information. The input of RNN includes the new input value at the current time and the output value of itself at the previous time. RNN is suitable for obtaining sequence features with correlation in time, such as speech recognition, channel coding and decoding and the like.

[0143] In the above model training process of 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 the specific form of the loss function is not limited. The model training process can be regarded as the following process: by adjusting part or all of the parameters of the model, the value of the loss function is less than the threshold value or meets the target requirement.

[0144] The model can also be referred to as an AI model, a rule or other names, etc. The AI model can be considered as a specific method to realize the AI function. The AI model represents the mapping relationship or function between the input and the output of the model. The AI function can include one or more of the following: data collection, model training (or model learning), model information publishing, model inference (or model reasoning, reasoning, or prediction, etc.), model monitoring or model verification, or inference result publishing, etc. The AI function can also be referred to as an AI (related) operation, or an AI related function.

[0145] The implementation process of the neural network will be described below with reference to the accompanying drawings.

[0146] 1. Fully connected neural network, also known as multilayer perceptron (MLP).

[0147] As shown in FIG. 2c, an MLP includes an input layer (left side), an output layer (right side), and multiple hidden layers (middle). Each layer of the MLP includes a plurality of nodes, referred to as neurons. The neurons of adjacent two layers are connected to each other.

[0148] Optionally, considering the neurons of adjacent two layers, the output h of the neuron of the next layer is the weighted sum of all the neurons x of the previous layer connected to it and is processed by an activation function, which can be expressed as: h = f (wx + b).

[0149] Where w is a weight matrix, b is a bias vector, and f is an activation function.

[0150] Further optionally, the output of the neural network can be recursively expressed as: y = f z (w z f z-1 (…)+b z ).

[0151] wherein z is an index of the neural network layer, z is greater than or equal to 1, and z is less than or equal to Z, wherein Z is the total number of layers of the neural network.

[0152] In other words, the neural network can be understood as a mapping relationship from a set of input data to a set of output data. Usually, the neural network is randomly initialized, and the process of obtaining this mapping relationship from random w and b with existing data is called training of the neural network.

[0153] Optionally, the specific manner of training is to evaluate the output result of the neural network by using a loss function.

[0154] As shown in FIG. 2d, the error can be back-propagated, and the neural network parameters (including w and b) can be iteratively optimized by the gradient descent method until the output of the loss function reaches a minimum value, i.e., the "better point (e.g., optimal point)" in FIG. 2d. It can be understood that the neural network parameters corresponding to the "better point (e.g., optimal point)" in FIG. 2d can be used as the neural network parameters in the trained AI model information.

[0155] Further optionally, the process of gradient descent can be represented as:

[0156] wherein θ is the parameter to be optimized (including w and b), L is the loss function, η is the learning rate, and controls the step size of gradient descent, represents the derivation operation, represents the derivative of L with respect to θ.

[0157] Further optionally, the process of back-propagation utilizes the chain rule of partial derivative.

[0158] As shown in FIG. 2e, the gradient of the parameters of the previous layer can be recursively calculated from the gradient of the parameters of the next layer, which can be expressed as:

[0159] wherein w ij is the weight of node j connected to node i, and s i is the input weighted sum on node i.

[0160] 2. Federated learning (FL).

[0161] The concept of federated learning effectively solves the difficulties faced by the current development of artificial intelligence. Under the premise of fully guaranteeing the privacy and security of user data, the learning task of the model is efficiently completed by promoting the cooperation of various edge devices and central servers.

[0162] As shown in FIG. 2f, the FL architecture is the most widely used training architecture in the current FL field, and the FedAvg algorithm is the basic algorithm of FL. The algorithm process of FedAvg is roughly as follows:

[0163] (1) The center end initializes the to-be-trained model and broadcasts it to all client ends.

[0164] (2) In the t-th round, the client end k based on the local data set performs E epochs of training on the received global model to obtain the local training result , which is reported to the center node. In the example shown in FIG. 2f, the local training results sent by distributed nodes n, k and m are denoted as G n , G k and G m , respectively.

[0165] (3) The center node collects the local training results from all (or part of) client ends. Assuming that the client end set uploading the local model in the t-th round is The center end obtains a new global model by weighted averaging with the sample number of the corresponding client end as the weight. The specific updating rule is After that, the center end broadcasts the latest version of the global model to all client ends for a new round of training.

[0166] (4) Repeat steps (2) and (3) until the model converges or the training round number reaches the upper limit.

[0167] Optionally, in addition to reporting the local model , the client end can also report the trained local gradient . The center node averages all the local gradients reported by the client ends and updates the global model according to the average gradient.

[0168] As can be seen, in the FL framework, the data set exists in the distributed node (such as the client end), that is, the distributed node collects the local data set and performs local training, and reports the local result (model or gradient) obtained by training to the center node. The center node itself can have no data set and can be responsible for fusing the training results of the distributed nodes to obtain a global model and issuing it to the distributed nodes.

[0169] 3. Decentralized learning.

[0170] As shown in FIG. 2g, it is a completely distributed system without a center node. The design goal f(x) of the decentralized learning system is generally the goal f i ​the mean of (x), i.e. where n is the number of distributed nodes, x is the parameter to be optimized, and in machine learning, x is the parameter of a machine learning (such as a neural network) model. Each node uses local data and local objective f i (x) calculates the local gradient Then it is sent to the neighbor nodes that are communicatively reachable. After receiving the gradient information sent by the neighbor nodes, any node can update the parameters x of the local model according to the following formula:

[0171] wherein, represents the parameters of the local model of the i-th node after the k+1 (k is a natural number) update, represents the parameters of the local model of the i-th node after the k update (if k is 0, it represents is the parameter of the local model of the i-th node that does not participate in the update), a k represents the tuning coefficient, N i is the neighbor node set of node i, |N i represents the number of elements in the neighbor node set of node i, i.e., the number of neighbor nodes of node i. Through the information interaction between nodes, the decentralized learning system will eventually learn a unified model.

[0172] The technical solution provided in the present application can be applied in a communication system (such as the system shown in FIG. 1a or FIG. 1b or FIG. 1c). In the communication system, the communication nodes generally have signal transceiving capability and computing capability. Taking a network device with computing capability as an example, the computing capability of the network device is mainly to provide computing power support for the signal transceiving capability (such as: to perform sending processing and receiving processing on the signal), so as to realize the communication task of the network device and other communication nodes.

[0173] With the development of communication technology, in the communication system, the communication device performs services in addition to traditional communication services, which can also include other new services, such as artificial intelligence (AI) services. Generally, a system capable of processing AI services, such as a communication system, can also be referred to as an AI system.

[0174] Currently, a communication device can be a participating node of an AI system, and the computing power of the communication device can be applied to a certain link of the AI system. Generally, the AI function introduced in the communication network needs to rely on a model to be implemented, and the communication device can process the input data through the model to obtain the output data. For example, taking a terminal device as the communication device, the AI use cases deployed in the terminal device can include CSI feedback enhancement, beam management enhancement, positioning enhancement, etc., and different use cases correspond to independent models, and the current situation does not involve mixed use cases. When the model is on the terminal device side, the network device can perform function level identification or model level identification. For function identification, the network device needs to know which AI functions the terminal device supports, for example, the above three AI use cases can be considered as three AI functions, and the network device does not need to know which model is used to implement each function. For model identification, the network device needs to know which AI models the terminal device has and the function corresponding to each model. The same function can have multiple models, and each model corresponds to different scenarios or conditions.

[0175] In addition, the way to improve the complexity of the model (for example, increasing the number of parameters of the model, increasing the number of neural network layers contained in the model, etc.) can effectively improve the performance of the output data. For example, a radio frequency map (RF MAP) model can be used to process relatively complex communication-related parameters. The input of the RF MAP model is environmental information or an environmental map, and the positions of two nodes in the environmental information, and the output of the model is the wireless channel parameters and wireless transmission parameters between the two nodes.

[0176] For example, the above wireless channel parameters include the power, delay, azimuth angle of arrival (AoA), azimuth angle of departure (AoD), zenith angle of departure (ZoD) on L (L is a positive integer, and here L is greater than an example) paths, which can be denoted as {(power1, delay1, AoA1, AoD1, ZoA1, ZoD1)…(power L , delay L , AoA L , AoD L , ZoA L , ZoD L )}.

[0177] For example, the wireless transmission parameter includes one or more of uplink and downlink MCS, uplink transmit power, uplink timing advance (TA), downlink transmission beam (Tx beam) indication, reference signal received power (RSRP), and uplink and downlink codebook.

[0178] In the data processing process of the RF MAP model, when a position in any environment is input, corresponding channel information and transmission information can be directly obtained without sending a reference channel for measurement. Obviously, the RF map is more complex than the above three use cases and can support multiple different types of input or output.

[0179] However, as the complexity of the model gradually increases, it is possible to cause the above data processing process to consume a relatively long time, thereby affecting the data processing efficiency. In the RF MAP model, because multiple types of input and output can be supported, the model corresponding to the use case is usually very large and the structure of the model is relatively complex. When inference is performed using the model corresponding to the use case, a large power consumption, a long inference delay, and a large amount of calculation resources are caused, and the requirement for the capability of the inference node is high. In addition, although the RF MAP model can support multiple inputs and outputs, in some cases, the inference node can only need part of the output result, for example, the terminal device only needs the wireless channel parameter and does not need the wireless transmission parameter. In this case, the complete model can not be inferred. Accordingly, in this case, the data processing delay is too long in the process in which the RF MAP model performs data inference based on the complete model, and some unnecessary overhead of power consumption and computing resource consumption of the communication device is caused, thereby affecting the data processing efficiency.

[0180] To solve the above problems, the present application provides a communication method and related devices, which will be described in detail below in conjunction with the accompanying drawings.

[0181] Please refer to FIG. 3, which is an implementation schematic diagram of a communication method provided by the present application. The method includes the following steps.

[0182] It should be noted that in the following, the first communication device and other communication devices (for example, the second communication device) in FIG. 3 are taken as an example to illustrate the execution subject of the interaction, but the present application does not limit the execution subject of the interaction. For example, the communication device can be a communication device (for example, a terminal device or a network device), or a chip, a baseband chip, a modem chip, an SoC chip (such as an SoC chip containing a modem core), a SIP chip, a communication module, a chip system, a processor, a logic module, or software in the communication device, etc.

[0183] As an example, the first communication device can be a terminal device and the second communication device can be a network device.

[0184] As another example, the first communication device can be a network device and the second communication device can be a terminal device.

[0185] As another example, the first communication device and the second communication device are both terminal devices, i.e., the scheme shown in FIG. 3 can be applied to a sidelink communication scenario.

[0186] S301. The first communication device determines a first sub-model. The first sub-model includes M model units, which are part of N model units, M is a positive integer, and N is greater than M. The N model units are used to determine a first model, which includes at least two sub-models, and the first sub-model is one of the at least two sub-models.

[0187] S302. The first communication device processes the first input data based on the first sub-model to obtain first output data.

[0188] In this application, a model can include an AI model, a neural network model, an AI neural network model, a machine learning model, or an AI processing model. Similarly, a sub-model can include an AI sub-model, a neural network sub-model, an AI neural network sub-model, a machine learning sub-model, or an AI processing sub-model.

[0189] In this application, in the process of processing input data based on a model or a sub-model to obtain output data by a communication device, the data processing can include one or more of reasoning, prediction, deduction, identification, and decision-making. For example, the input data can be reasoning data, and the output data can be reasoning output data. For example, in the above process, the first communication device can process the first input data based on the first sub-model to obtain the first output data, including one or more of reasoning, prediction, deduction, identification, and decision-making.

[0190] It should be understood that a model can be understood as a whole model, any sub-model contained in the one model can be a partial model, and the any sub-model can include one or more model components. In other words, the model involved in this application can be implemented in three granularities, including a whole model, a partial model, and a model component. In the above scheme, the whole model is denoted as a model (e.g., the first model), the partial model is denoted as a sub-model (e.g., the at least two sub-models included in the first model, the first sub-model, etc.), and the model component is denoted as a model unit (e.g., the N model units, the M model units, etc.). These terms can be replaced by other descriptions.

[0191] As an example, the whole model can be a model, the part model can be a sub-model group, and the model component can be a sub-model. For example, in the above scheme, the sub-model can be replaced by a sub-model group (e.g., at least two sub-models can be replaced by at least two sub-model groups, and the first sub-model can be replaced by a first sub-model group), and the model unit can be replaced by a sub-model (e.g., N model units can be replaced by N sub-models, and M model units can be replaced by M sub-models).

[0192] As another example, the whole model can be a joint model, the part model can be a model, and the model component can be a sub-model. For example, in the above scheme, the model can be replaced by a joint model (e.g., the first model can be replaced by a first joint model), the sub-model can be replaced by a model (e.g., at least two sub-models can be replaced by at least two models, and the first sub-model can be replaced by a first model), and the model unit can be replaced by a sub-model (e.g., N model units can be replaced by N sub-models, and M model units can be replaced by M sub-models).

[0193] It should be noted that, in the at least two sub-models included in the first model, the one or more model units and the connection order included in any sub-model are different from those included in other sub-models.

[0194] For example, the one or more model units included in any sub-model are different from those included in other sub-models.

[0195] For another example, the one or more model units included in a certain sub-model are the same as those included in another sub-model. Moreover, the connection order between the one or more model units included in the certain sub-model is different from that between the one or more model units included in the another sub-model.

[0196] In a possible implementation, the model information of any sub-model (or at least one sub-model) is pre-configured (e.g., the related parameters involved in Tables 2, 3, and 4 below can be pre-configured), which can reduce implementation complexity and communication overhead.

[0197] Optionally, the model information of any sub-model (or at least one sub-model) can be configured by a network device, so as to improve the flexibility of the scheme implementation.

[0198] Optionally, in the at least two sub-models included in the first model, the model information of any sub-model includes at least one of: an index corresponding to the any sub-model, a use case corresponding to the any sub-model, an AI feature corresponding to the any sub-model, a function corresponding to the any sub-model, a task corresponding to the any sub-model, a data type of input data of the any sub-model, a data type of output data of the any sub-model, an identifier of one or more model units included in the any sub-model, a connection relationship between the one or more model units included in the any sub-model, or a target performance corresponding to the any sub-model. For details, refer to Table 2, Table 3, Table 4, and the related implementation process below.

[0199] In a possible implementation, in the at least two sub-models, the target performances corresponding to different sub-models are different (for details, refer to Table 4 and the related implementation process below). Specifically, in the at least two sub-models included in the first model, the target performances corresponding to different sub-models are different, so that the first communication apparatus (or the provider of the first information above) can cause the first communication apparatus to perform data processing on the corresponding sub-model based on the performance requirement or the performance indication, so as to meet the performance requirement or the performance indication.

[0200] Based on the scheme shown in FIG. 3, after the first communication apparatus determines the first sub-model in the at least two sub-models included in the first model in step S301, the first communication apparatus can process the first input data based on the first sub-model to obtain the first output data in step S302. Thus, compared with the process in which the communication apparatus processes the input data based on the complete model (i.e., the first model) to obtain the output data, in the scheme above, the first communication apparatus processes the input data based on one of the sub-models included in the model to obtain the output data, which can reduce the complexity of data processing and also reduce the data processing delay, thereby improving the data processing efficiency.

[0201] In addition, the first model is determined by N model units, the first sub-model is determined by M model units in the N model units, and M is less than N. In this way, in the process in which the first communication apparatus processes the first input data to obtain the first output data, the first communication apparatus can not need to involve other N-M model units, which can reduce the computing power consumption and power consumption of the communication apparatus.

[0202] To facilitate understanding of the implementation process shown in FIG. 3, some implementation examples will be provided below, taking the overall model as the model (for example, the first model), the partial model as the sub-model (for example, the at least two sub-models included in the first model, the first sub-model, etc.), and the model component as the model unit (for example, the N model units, the M model units, etc.) as examples for description.

[0203] As an example, as shown in FIG. 4a, the first model can include 7 (N=7) model units in the figure:

[0204] Model unit 1-1, associated with indoor environment, or for processing data related to indoor environment;

[0205] Model unit 1-2, associated with outdoor environment, or for processing data related to outdoor environment;

[0206] Model unit 2, associated with wireless channel, or for processing data related to wireless channel;

[0207] Model unit 3-1, associated with MCS, or for processing data related to MCS;

[0208] Model unit 3-2, associated with power, or for processing data related to power;

[0209] Model unit 3-3, associated with beam, or for processing data related to beam;

[0210] Model unit 3-4, associated with codebook, or for processing data related to codebook.

[0211] Through the model shown in FIG. 4a, the input data can include the location information and / or the environment information of the communication device, and the output data can include some or all of the following 16 pieces of information of the communication device: MCS of indoor environment, power of indoor environment, beam of indoor environment, codebook of indoor environment, MCS of wireless channel of indoor environment, power of wireless channel of indoor environment, beam of wireless channel of indoor environment, codebook of wireless channel of indoor environment, MCS of outdoor environment, power of outdoor environment, beam of outdoor environment, codebook of outdoor environment, MCS of wireless channel of outdoor environment, power of wireless channel of outdoor environment, beam of wireless channel of outdoor environment, codebook of wireless channel of outdoor environment.

[0212] In other words, the first model shown in FIG. 4a supports 16 use cases / AI features / tasks / functions, respectively corresponding to the 16 pieces of information contained in the above output data. According to the conventional data processing mode, in the case that the first communication device needs or other devices indicate one or more of the 16 use cases / AI features / tasks / functions, the first communication device needs to participate in the data processing process through all the 7 model units shown in FIG. 4a, in order to obtain the corresponding output data. For example, in the case that the first communication device needs or other devices indicate the MCS of indoor channel, the first communication device needs to completely execute the 7 model units shown in FIG. 4a, in order to obtain the MCS of indoor channel.

[0213] From the above process, in the case that the first communication device needs or other devices indicate part of the 16 use cases / AI characteristics / tasks / functions, through the data processing process, since the output data corresponding to other use cases / AI characteristics / tasks / functions is not required by the first communication device itself or indicated by other devices, it is possible that the output data corresponding to the other use cases / AI characteristics / tasks / functions is not used, thereby causing unnecessary power consumption and computing power consumption.

[0214] In the scheme shown in FIG. 3, the first model containing 7 (N = 7) model units can correspond to different sub-models through Table 2 as follows.

[0215] Table 2

[0216] As can be seen from Table 2, the first model containing 7 (N = 7) model units (i.e., model unit 1-1, model unit 1-2, model unit 2, model unit 3-1, model unit 3-2, model unit 3-3 and model unit 3-4 in FIG. 4a) can correspond to the following four sub-models (i.e., here taking at least two sub-models contained in the first model as four sub-models as an example).

[0217] The number of model units contained in sub-model 1 is 3, which are model unit 1-1, model unit 2 and model unit 3-1.

[0218] The number of model units contained in sub-model 2 is 3, which are model unit 1-2, model unit 2 and model unit 3-2.

[0219] The number of model units contained in sub-model 3 is 3, which are model unit 1-2, model unit 2 and model unit 3-3.

[0220] The number of model units contained in sub-model 4 is 3, which are model unit 1-2, model unit 2 and model unit 3-4.

[0221] It should be understood that the 7 (N = 7) model units shown in FIG. 4a can obtain other sub-models through other combination manners in addition to the above sub-model 1 to sub-model 4, and the four sub-models provided here are only some possible implementation examples.

[0222] According to the above implementation, the first sub-model can be the sub-model 1, the sub-model 2, the sub-model 3, or the sub-model 4. For example, in the case that the first communication device has a demand or other devices indicate the MCS of the indoor channel, the first communication device can obtain the MCS of the indoor channel through the 3 model units included in the sub-model 1 (i.e., the first sub-model described above can be the sub-model 1, and M = 3), so that the first communication device does not need to completely execute the 7 model units shown in FIG. 4a, and the power consumption and the computing power consumption can be reduced, and the data processing efficiency is improved.

[0223] As another example, as shown in FIG. 4b, the first model can include 6 (N = 6) model units in the figure:

[0224] The model unit 1-1 is associated with the indoor environment, or is used for processing data related to the indoor environment;

[0225] The model unit 1-2 is associated with the outdoor environment, or is used for processing data related to the outdoor environment;

[0226] The model unit 2-1 is associated with the MCS, or is used for processing data related to the MCS;

[0227] The model unit 2-2 is associated with the power, or is used for processing data related to the power;

[0228] The model unit 2-3 is associated with the beam, or is used for processing data related to the beam;

[0229] The model unit 2-4 is associated with the codebook, or is used for processing data related to the codebook.

[0230] According to the model shown in FIG. 4b, the input data can include the position information and / or the environment information of the communication device, and the output data can include the following 8 pieces of information of the communication device: the MCS of the indoor environment, the power of the indoor environment, the beam of the indoor environment, the codebook of the indoor environment, the MCS of the outdoor environment, the power of the outdoor environment, the beam of the outdoor environment, and the codebook of the outdoor environment.

[0231] In other words, the first model shown in FIG. 4b supports 8 use cases / AI features / tasks / functions, which respectively correspond to the 8 pieces of information included in the output data. According to the conventional data processing mode, in the case that the first communication device has a demand or other devices indicate one or more of the 8 use cases / AI features / tasks / functions, the first communication device needs to participate in the data processing process through the 6 model units shown in FIG. 4b, and then the corresponding output data can be obtained. For example, in the case that the first communication device has a demand or other devices indicate the beam of the outdoor channel, the first communication device needs to completely execute the 6 model units shown in FIG. 4b, and then the beam of the outdoor channel can be obtained.

[0232] From the above process, in the case that the first communication device needs or other devices indicate part of the eight use cases / AI characteristics / tasks / functions, through the data processing process, since the output data corresponding to other use cases / AI characteristics / tasks / functions is not required by the first communication device or indicated by other devices, it is possible that the output data corresponding to the other use cases / AI characteristics / tasks / functions is not used, thereby causing unnecessary power consumption and computing power consumption.

[0233] In the scheme shown in FIG. 3, the first model including 6 (N = 6) model units (i.e., model unit 1-1, model unit 1-2, model unit 2-1, model unit 2-2, model unit 2-3, and model unit 2-4 in FIG. 4b) can correspond to the following four different sub-models (i.e., taking at least two sub-models included in the first model as four sub-models as an example) respectively through Table 3 as follows.

[0234] Table 3

[0235] As can be seen from Table 3, the 6 (N = 6) model units included in the first model can correspond to the following sub-models.

[0236] The number of model units included in the sub-model 1 is 2, which are model unit 1-1 and model unit 2-1.

[0237] The number of model units included in the sub-model 2 is 2, which are model unit 1-2 and model unit 2-2.

[0238] The number of model units included in the sub-model 3 is 2, which are model unit 1-2 and model unit 2-3.

[0239] The number of model units included in the sub-model 4 is 2, which are model unit 1-2 and model unit 2-4.

[0240] It should be understood that the 6 (N = 6) model units shown in FIG. 4b can obtain other sub-models through other combination manners in addition to the above-mentioned sub-model 1 to sub-model 4, and the four sub-models provided herein are only some possible implementation examples.

[0241] It can be understood from the above implementation that the first sub-model can be the above-mentioned sub-model 1, sub-model 2, sub-model 3, or sub-model 4. For example, in the case that the first communication device needs or other devices indicate the MCS of the outdoor channel, the first communication device can obtain the beam of the outdoor channel through the 2 model units included in the sub-model 3 (i.e., the first sub-model described above can be the sub-model 3, and M = 2), so that the first communication device does not need to completely perform the 6 model units shown in FIG. 4b, and the power consumption and computing power consumption can be reduced, and the data processing efficiency is improved.

[0242] As another example, as shown in FIG. 4c, the first model can include 8 (N = 8) model units in the figure: encoding unit 0, encoding unit 1, encoding unit 2, encoding unit 3, decoding unit 0, decoding unit 1, decoding unit 2, and decoding unit 3.

[0243] It should be understood that the encoder can be used to encode the input data to obtain feature information, which can be understood as a feature extraction and abstraction of the input data; and the decoder can be used to decode the feature information to obtain the output data. Generally, the encoder and the decoder can be used in combination to form a model. For example, in the case that the model is used for a machine translation task, the encoder is responsible for understanding the sentence in the source language, and the decoder generates the translation in the target language according to the representation generated by the encoder. Such a model structure can learn the complex mapping relationship between the input and the output, without the need for explicit design of feature engineering, and can enable the model to automatically learn the optimal feature representation and generation strategy.

[0244] Optionally, each encoding unit can include one or more attention modules, and the number of attention modules included in different encoding units can be the same or different, which is not limited here. The attention module plays a crucial role in the AI field, especially when dealing with complex data with inherent structure, such as natural language, images, and videos. The attention mechanism allows the model to focus on the most relevant or most important part when processing the input, thereby improving the efficiency and accuracy of the model.

[0245] Similarly, each decoding unit can also include one or more attention modules, and the number of attention modules included in different decoding units can be the same or different, which is not limited here. In FIG. 4c, taking the decoding unit 3 including k attention modules as an example, k is a positive integer.

[0246] In the model shown in FIG. 4c, one or more encoding units included in the encoder and one or more encoding units included in the decoder can be used to implement part or all of CSI prediction, beam prediction, positioning, etc. use cases / AI features / tasks / functions. Taking the model shown in FIG. 4c as an example for implementing CSI prediction, the input data can include channel data of a lower dimension, and the output data can include channel data of a higher dimension. For example, the dimension can be determined by one or more of the number of antenna ports of the sending end, the number of antenna ports of the receiving end, and the number of frequency domain units.

[0247] Generally, the more the number of encoding units included in the encoder (and / or the number of decoding units included in the decoder), the higher the target performance of the model in performing CSI prediction. For example, the more the number of encoding units, the higher the target performance of the model in performing CSI prediction; conversely, the less the number of encoding units, the lower the target performance of the model in performing CSI prediction. The implementation process will be described below in conjunction with some examples.

[0248] For example, in FIG. 4c, when the number of encoding units included in the encoder and the number of decoding units included in the decoder are both 4, the target performance (e.g., CSI prediction accuracy) of CSI prediction is 99%.

[0249] For another example, in FIG. 4c, when the number of encoding units included in the encoder and the number of decoding units included in the decoder are both 3, the target performance (e.g., CSI prediction accuracy) of CSI prediction is 95%.

[0250] For another example, in FIG. 4c, when the number of encoding units included in the encoder and the number of decoding units included in the decoder are both 2, the target performance (e.g., CSI prediction accuracy) of CSI prediction is 90%.

[0251] For another example, in FIG. 4c, when the number of encoding units included in the encoder and the number of decoding units included in the decoder are both 1, the target performance (e.g., CSI prediction accuracy) of CSI prediction is 85%.

[0252] According to the conventional data processing mode, the first communication device can deploy models with different target performances in advance. In the case that the first communication device needs to obtain output data according to a certain target performance indicated by itself or other devices, the first communication device needs to participate in the data processing process through the model corresponding to the target performance, so as to obtain the corresponding output data. According to the model shown in FIG. 4c, the first communication device can obtain the output with the target performance (for example, the CSI prediction accuracy) of 99% of the CSI prediction through the processing of the four encoding units and the four decoding units included in the model. For example, in the case that the first communication device needs to obtain the target performance (for example, the CSI prediction accuracy) of 90% of the CSI prediction indicated by itself or other devices, if the first communication device does not deploy the corresponding model, the first communication device needs to completely execute the eight model units shown in FIG. 4c, so as to obtain the output with the target performance (for example, the CSI prediction accuracy) of 99% of the CSI prediction, while the first communication device may only need to obtain the target performance of 90% at this time.

[0253] In the scheme shown in FIG. 3, the first model including eight (N=8) model units can correspond to different sub-models through Table 4, respectively.

[0254] Table 4

[0255] As shown in Table 4, the eight (N=8) model units (that is, model unit 1-1, model unit 1-2, model unit 2, model unit 3-1, model unit 3-2, model unit 3-3 and model unit 3-4 in FIG. 4c) included in the first model can correspond to the following four sub-models (for example, at least two sub-models included in the first model are taken as four sub-models here).

[0256] The number of model units included in the sub-model 1 is eight, which are encoding modules 0 / 1 / 2 / 3 and decoding modules 0 / 1 / 2 / 3, respectively. The corresponding target performances include at least one of the accuracy of 99%, the relative throughput of 99% and the relative RSRP of 0.5 dB.

[0257] The number of model units included in the sub-model 2 is six, which are encoding modules 0 / 1 / 2 and decoding modules 0 / 1 / 2, respectively. The corresponding target performances include at least one of the accuracy of 95%, the relative throughput of 95% and the relative RSRP of 1 dB.

[0258] The number of model units included in the sub-model 3 is four, which are encoding modules 0 / 1 / 2 / 3 and decoding modules 0 / 1 / 2 / 3, respectively. The corresponding target performances include at least one of the accuracy of 90%, the relative throughput of 90% and the relative RSRP of 2 dB.

[0259] The sub-model 4 includes 2 model units, which are the encoding modules 0 / 1 / 2 / 3 and the decoding modules 0 / 1 / 2 / 3, and the corresponding target performance includes at least one of the accuracy of 80%, the relative throughput of 80%, and the relative RSRP of 3 dB.

[0260] It should be understood that, in addition to the above sub-model 1 to sub-model 4, the 8 (N=8) model units shown in FIG. 4c can also obtain other sub-models through other combinations. The four sub-models provided herein are only some possible implementation examples.

[0261] It should be understood that, in the above Table 4, the target performance corresponding to different sub-models (for example, the accuracy, relative throughput, and relative RSRP of any two sub-models) can be different, and accordingly, different target performances can be used to uniquely indicate a certain sub-model in Table 4. For this purpose, the target performance can also serve as an identifier / index of different sub-models.

[0262] It can be known from the above implementation that the first sub-model can be the above sub-model 1, sub-model 2, sub-model 3, or sub-model 4. For example, in the case that the first communication device requires or other devices indicate that the accuracy of CSI prediction is 90%, the first communication device can obtain the output of the accuracy of CSI prediction of 90% through the 4 model units included in the sub-model 3 (that is, the first sub-model described above can be the sub-model 3, and M=4), so that the first communication device does not need to completely perform the 8 model units shown in FIG. 4c, and the power consumption and computing power consumption can be reduced, and the data processing efficiency is improved.

[0263] In a possible implementation, the method shown in FIG. 3 further includes:

[0264] S300. The first communication device sends first information, and correspondingly, the second communication device receives the first information. Or, the second communication device sends first information, and correspondingly, the first communication device receives the first information. The first information is used to indicate the first sub-model or model information of the first sub-model. Through the above manner, the first communication device can send the first information, so that the receiver of the first information can determine the sub-model or model information of the sub-model used by the first communication device for data processing based on the first information, and then the receiver can perform model management (for example, one or more of model inference, model scheduling, model updating, model switching, or function fallback) on the sub-model used by the first communication device for data processing based on the first information.

[0265] Or, the method shown in FIG. 3 further includes:

[0266] S300. The second communication device sends the first information, and correspondingly, the first communication device receives the first information. Wherein, the first information is used to indicate the first sub-model or the model information of the first sub-model. Through the above manner, the first communication device can receive the first information, so that the first communication device can determine the sub-model or the model information of the sub-model for the first communication device to perform data processing based on the first information, and further make the first communication device perform data processing on the model deployed by the first communication device based on the indication of the other communication device.

[0267] As an example, as known from the foregoing, the model information of any sub-model can be implemented in various manners, and correspondingly, the first information can indicate the first sub-model or the model information of the first sub-model in various manners.

[0268] For example, the identifiers / indexes of different sub-models can be different, and therefore, the first information can include the identifier / index of the first sub-model. Taking the first sub-model indicated by the first information as the sub-model 3 in Table 2 in the foregoing as an example, the first information can include the index “3”, so that the receiver of the first information can determine the first sub-model based on the value of the index.

[0269] For another example, the target performances of different sub-models can be different, and therefore, the first information can include the indication information of the target performance of the first sub-model. Taking the first sub-model indicated by the first information as the sub-model 3 in Table 4 in the foregoing as an example, the first information can include at least one of the accuracy of 90%, the relative throughput of 90%, and the relative RSRP of 2 dB, so that the receiver of the first information can determine the first sub-model based on the value of the performance. Correspondingly, the receiver of the first information can also determine the target performance required / indicated by the sender of the first information based on the indication information of the target performance included in the first information, and subsequently obtain the output data corresponding to the target performance through the data processing process of the first sub-model to meet the requirement / indication.

[0270] For another example, the model information of different sub-models can be different, and therefore, the first information can include the model information of the first sub-model. Taking the first sub-model indicated by the first information as the sub-model 2 in Table 3 in the foregoing as an example, the first information can include at least one item of information corresponding to “sub-model 2” in Table 3 (i.e., at least one of the indication information indicating that the use case / AI feature / function / task is a beam for an outdoor environment, the indication information indicating that the input data of the model includes outdoor channel information, the indication information indicating that the output data of the model is a beam, the indication information indicating that the activated model units are model units 1-2 and model units 2-3, and the indication information indicating the connection relationship between the two model units), so that the receiver of the first information can determine the first sub-model based on the at least one item.

[0271] In a possible implementation, as shown in the example of FIG. 4d, the method shown in FIG. 3 further includes:

[0272] In step A, the second communication device sends second information, and the first communication device receives the second information. The second information is used to indicate the first input data, and the first input data is used to determine the first sub-model. Specifically, the first communication device can receive the second information indicating the first input data, and can determine the first sub-model from the at least two sub-models based on the first input data. In other words, the first input data can be used to determine one or more sub-models from the at least two sub-models, and the first communication device can select the first sub-model for processing the first input data from the one or more sub-models.

[0273] Optionally, the first communication device can randomly select the first sub-model from the one or more sub-models. Alternatively, the first communication device can select the first sub-model from the one or more sub-models based on its own state information and / or requirements, so that the process of processing the first input data by the first communication device can meet the state indicated by the state information and / or the requirements.

[0274] For example, the state information can include one or more of the following parameters: geographical position, moving speed, signal coverage strength, signal to interference plus noise ratio (SINR) change, channel response change, data throughput change, and the like. For another example, the AI state information can include one or more of the following parameters: (current) model performance, data distribution of model input data, data distribution of model output data, and the like.

[0275] Optionally, the second information can include the first input data, or the second information can include data collection configuration information of the first input data. For example, the first input data can include a reference signal, and the data collection configuration information of the first input data can be configuration information of the reference signal.

[0276] Optionally, for the first communication device, the first input data can be collected or obtained by the first communication device itself, so that the input data of the model can be determined without the indication of other communication devices, and the overhead can be reduced.

[0277] In a possible implementation, as shown in the example of FIG. 4d, the method shown in FIG. 3 further includes:

[0278] Step B. The second communication device sends third information, and the first communication device receives the third information. The third information is used to indicate the first model. Alternatively, the first communication device obtains the first model based on the training data. Thus, the first communication device can obtain the first model by any of the above manners, so as to improve the flexibility of the implementation of the scheme.

[0279] Optionally, in the case that the first communication device obtains the first model locally based on the training data, the first communication device can send indication information to the second communication device. The indication information is used to indicate the first model. For example, in the case that the first model obtained by the first communication device locally includes the first model including 7 model units shown in Table 2, the first communication device can indicate part or all of the information of Table 2 to the second communication device, so that the second communication device can know the first model currently deployed by the first communication device, so as to facilitate the second communication device to subsequently perform model management (such as one or more of model inference, model scheduling, model updating, model switching, or function fallback) on the first model deployed by the first communication device.

[0280] In a possible implementation, as shown in the example of FIG. 4d, the method shown in FIG. 3 further includes:

[0281] Step C. The second communication device sends fourth information, and the first communication device receives the fourth information. The fourth information is used to indicate the second sub-model or model information of the second sub-model. The first communication device processes the first input data based on the first sub-model to obtain first output data, including: in the case that the first communication device determines that the state information does not meet the second sub-model, the first communication device processes the first input data based on the first sub-model to obtain the first output data. Specifically, the first communication device can receive the fourth information indicating the second sub-model or the model information of the second sub-model. In the case that the first communication device determines that the state information does not meet the second sub-model, the first communication device can perform data processing based on other sub-models (for example, the first sub-model), so as to improve the data processing efficiency and reduce the latency.

[0282] Similarly, the state information can include the communication state and / or the AI state of the first communication device, which can be referred to the foregoing description.

[0283] Optionally, after step C, the method shown in FIG. 3 further includes:

[0284] Step D. The first communication apparatus sends fifth information, and the second communication apparatus receives the fifth information. The fifth information is used to indicate that the state information of the first communication apparatus does not satisfy the second sub-model. Specifically, the first communication apparatus can send the fifth information, so that the receiver of the fifth information can explicitly know that the state information of the first communication apparatus does not satisfy the second sub-model, and explicitly know that the first communication apparatus performs data processing based on other sub-models (for example, the first sub-model).

[0285] Optionally, in the case where it is determined that the state information of the first communication apparatus does not satisfy the second sub-model, the first communication apparatus can randomly select or select one of other sub-models based on the state information as the sub-model for data processing, other than the second sub-model. Alternatively, after the first communication apparatus sends the fifth information, the receiver of the fifth information can send first information (which can refer to the foregoing implementation) to the first communication apparatus based on the fifth information, so that the first communication apparatus explicitly knows to take the first sub-model as the sub-model for data processing based on the first information.

[0286] In a possible implementation, as shown in the example of FIG. 4d, the method shown in FIG. 3 further includes:

[0287] Step E. The first communication apparatus sends sixth information, and the second communication apparatus receives the sixth information. The sixth information is used to indicate the first output data. Thus, the receiver of the sixth information can obtain the first output data obtained by the first communication apparatus through data processing, so as to realize feedback / indication of model output data. Similarly, the receiver of the sixth information can also perform model management (for example, model inference, model scheduling, model updating, model switching, or function fallback, etc.) on the model deployed by the first communication apparatus based on the first output data.

[0288] Optionally, the first communication apparatus can use the first output data locally, that is, the first communication apparatus can also not send the sixth information.

[0289] Referring to FIG. 5, an embodiment of the present application provides a communication apparatus 500, which can implement the functions of the first communication apparatus (or the second communication apparatus) in the foregoing method embodiments, and thus can also achieve the beneficial effects possessed by the foregoing method embodiments. In the embodiment of the present application, the communication apparatus 500 can be the first communication apparatus (or the second communication apparatus), or an integrated circuit or element inside the first communication apparatus (or the second communication apparatus), for example, a chip, a baseband chip, a modem chip, an SoC chip (such as an SoC chip containing a modem core), a SIP chip, a communication module, a chip system, a processor, etc.

[0290] It should be noted that the transceiver unit 502 can include a transmitting unit and a receiving unit, which are respectively used for performing transmitting and receiving.

[0291] In a possible implementation, when the apparatus 500 is configured to perform the method performed by the first communication device in FIG. 3 and related embodiments, the apparatus 500 includes a processing unit 501; the processing unit 501 is configured to determine a first sub-model, the first sub-model including M model units, the M model units being part of N model units, M being a positive integer, and N being greater than M; wherein the N model units are used to determine a first model, the first model including at least two sub-models, the first sub-model being one of the at least two sub-models; and the processing unit 501 is further configured to process first input data based on the first sub-model to obtain first output data. Optionally, the processing unit 501 is configured to receive or send first information through the transceiver unit 502.

[0292] In a possible implementation, when the apparatus 500 is configured to perform the method performed by the second communication device in FIG. 3 and related embodiments, the apparatus 500 includes a processing unit 501 and a transceiver unit 502; the processing unit 501 is configured to determine first information, the first information being used to indicate a first sub-model or model information of the first sub-model; wherein the first sub-model includes M model units, the M model units being part of N model units, M being a positive integer, and N being greater than M; the N model units are used to determine a first model, the first model including at least two sub-models, the first sub-model being one of the at least two sub-models; and the transceiver unit 502 is configured to send the first information.

[0293] In a possible implementation, when the apparatus 500 is configured to perform the method performed by the second communication device in FIG. 3 and related embodiments, the apparatus 500 includes a processing unit 501 and a transceiver unit 502; the transceiver unit 502 is configured to receive first information, the first information being used to indicate a first sub-model or model information of the first sub-model; wherein the first sub-model includes M model units, the M model units being part of N model units, M being a positive integer, and N being greater than M; the N model units are used to determine a first model, the first model including at least two sub-models, the first sub-model being one of the at least two sub-models; and the processing unit 501 is configured to perform model management on the first sub-model based on the first information.

[0294] In a possible design, when the communication apparatus 500 is a communication module in a terminal device or terminal, the function of the processing unit 501 can be implemented by one or more processors. Specifically, the processor can include a modem chip, a SoC chip (such as a SoC chip including a modem core), or a SIP chip. The function of the transceiver unit 502 can be implemented by a transceiver circuit.

[0295] In a possible design, when the communication apparatus 500 is a circuit or chip responsible for communication functions in a terminal, such as a modem chip or a SoC chip or a SoC chip including a modem core or a SIP chip, the function of the processing unit 501 can be implemented by circuitry including one or more processors or processor cores in the chip. The function of the transceiver unit 502 can be implemented by an interface circuit or data transceiver circuit on the chip.

[0296] It should be noted that the information processing process and the like of the units of the communication apparatus 500 described above can be specifically refer to the descriptions in the method embodiments described above, and will not be repeated here.

[0297] Please refer to FIG. 6, which is another schematic structural diagram of a communication apparatus 600 provided in the present application, and the communication apparatus 600 includes a logic circuit 601 and an input / output interface 602. The communication apparatus 600 can be a chip or an integrated circuit.

[0298] The transceiver unit 502 shown in FIG. 5 can be a communication interface, which can be the input / output interface 602 in FIG. 6. The input / output interface 602 can include an input interface and an output interface. Alternatively, the communication interface can be a transceiver circuit, which can include an input interface circuit and an output interface circuit.

[0299] In a possible implementation, when the apparatus 600 is configured to perform the method performed by the first communication apparatus in FIG. 3 and related embodiments, the logic circuit 601 is configured to determine a first sub-model, the first sub-model including M model units, the M model units being part of N model units, M being a positive integer, and N being greater than M; the N model units being used to determine a first model, the first model including at least two sub-models, the first sub-model being one of the at least two sub-models; and the logic circuit 601 is further configured to process first input data based on the first sub-model to obtain first output data.

[0300] In a possible implementation, when the apparatus 600 is configured to perform the method performed by the second communication device in FIG. 3 and related embodiments, the logic circuit 601 is configured to determine first information, the first information being used to indicate a first sub-model or model information of the first sub-model, wherein the first sub-model includes M model units, the M model units being part of N model units, M being a positive integer, and N being greater than M, the N model units being used to determine a first model, the first model including at least two sub-models, the first sub-model being one of the at least two sub-models, and the input and output interface 602 is configured to send the first information.

[0301] In a possible implementation, when the apparatus 600 is configured to perform the method performed by the second communication device in FIG. 3 and related embodiments, the input and output interface 602 is configured to receive first information, the first information being used to indicate a first sub-model or model information of the first sub-model, wherein the first sub-model includes M model units, the M model units being part of N model units, M being a positive integer, and N being greater than M, the N model units being used to determine a first model, the first model including at least two sub-models, the first sub-model being one of the at least two sub-models, and the logic circuit 601 is configured to perform model management on the first sub-model based on the first information.

[0302] The logic circuit 601 and the input and output interface 602 can also perform other steps of the first communication device or the second communication device in any of the embodiments and achieve the corresponding beneficial effects, which are not described here.

[0303] In a possible implementation, the processing unit 501 shown in FIG. 5 can be the logic circuit 601 in FIG. 6.

[0304] Optionally, the logic circuit 601 can be a processing apparatus, and the functions of the processing apparatus can be partially or entirely implemented through software.

[0305] Optionally, the processing apparatus can include a memory and a processor, wherein the memory is configured to store a computer program, and the processor is configured to read and execute the computer program stored in the memory to perform the corresponding processing and / or steps in any of the method embodiments.

[0306] Optionally, the processing apparatus can include only the processor. The memory for storing the computer program is located outside the processing apparatus, and the processor is connected with the memory through a circuit / wire to read and execute the computer program stored in the memory. The memory and the processor can be integrated together or can be physically independent of each other.

[0307] Optionally, the processing device can be one or more chips, or one or more integrated circuits. For example, the processing device can be one or more field-programmable gate arrays (FPGA), application specific integrated circuits (ASIC), system on chips (SoC), central processing units (CPU), network processors (NP), digital signal processors (DSP), micro controller units (MCU), programmable logic devices (PLD) or other integrated chips, or any combination of the above chips or processors, etc.

[0308] Referring to FIG. 7, a communication device 700 involved in the above embodiments provided by the embodiments of the present application is shown, which can be the communication device as the terminal device in the above embodiments, and the example shown in FIG. 7 is implemented by the terminal device (or components in the terminal device).

[0309] Optionally, the communication device 700 can include but is not limited to at least one processor 701 and a communication port 702.

[0310] Optionally, the transceiver unit 502 shown in FIG. 5 can be a communication interface, which can be the communication port 702 in FIG. 7, and the communication port 702 can include an input interface and an output interface. Alternatively, the communication port 702 can also be a transceiver circuit, which can include an input interface circuit and an output interface circuit.

[0311] Further optionally, the device can further include at least one of a memory 703 and a bus 704, and in the embodiments of the present application, the at least one processor 701 is configured to control and process the actions of the communication device 700.

[0312] Further, the processor 701 can be a central processing unit, a general purpose processor, a digital signal processor, an application specific integrated circuit, a field programmable gate array, or other programmable logic device, transistor logic, hardware component, or any combination thereof. It can implement or execute various example logical blocks, modules, and circuits described in connection with the disclosure. The processor can also be a combination of computing functionality, such as a combination of one or more microprocessors, a combination of a digital signal processor and a microprocessor, or the like. For the sake of brevity and conciseness, the specific processes performed by the system, apparatus, and units described above can be referred to the corresponding processes in the method embodiments described above, and will not be described herein again.

[0313] It should be noted that the communication apparatus 700 shown in FIG. 7 can be specifically 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. The specific implementation of the communication apparatus shown in FIG. 7 can be referred to the description in the foregoing method embodiments, and will not be described herein again.

[0314] Please refer to FIG. 8, which is a structural schematic diagram of a communication apparatus 800 provided by an embodiment of the present application, and the communication apparatus 800 can be specifically the communication apparatus as the network device in the foregoing embodiments. The example shown in FIG. 8 is implemented by the network device (or components in the network device), and the structure of the communication apparatus can be referred to the structure shown in FIG. 8.

[0315] The communication apparatus 800 includes at least one processor 811 and at least one network interface 814. Further optionally, the communication apparatus further includes at least one memory 812, at least one transceiver 813, and one or more antennas 815. The processor 811, the memory 812, the transceiver 813, and the network interface 814 are connected, for example, through a bus. In the embodiments of the present application, the connection can include various interfaces, transmission lines, or buses, etc., and the embodiments of the present application do not limit the same. The antenna 815 is connected to the transceiver 813. The network interface 814 is used to enable the communication apparatus to communicate with other communication devices through a communication link. For example, the network interface 814 can include a network interface between the communication apparatus and the core network device, such as an S1 interface. The network interface can include a network interface between the communication apparatus and other communication apparatuses (such as other network devices or core network devices), such as an X2 or Xn interface.

[0316] The transceiver unit 502 shown in FIG. 5 can be a communication interface, which can be the network interface 814 in FIG. 8. The network interface 814 can include an input interface and an output interface. Alternatively, the network interface 814 can be a transceiver circuit, which can include an input interface circuit and an output interface circuit.

[0317] The processor 811 is mainly used for processing communication protocols and communication data, and controlling the whole communication device, executing software programs, processing data of the software programs, for example, for supporting the communication device to perform the actions described in the embodiments. The communication device can include a baseband processor mainly used for processing communication protocols and communication data, and a central processor mainly used for controlling the whole terminal device, executing software programs, and processing data of the software programs. The processor 811 in FIG. 8 can integrate the functions of the baseband processor and the central processor, and those skilled in the art can understand that the baseband processor and the central processor can also be independent processors interconnected by a bus or the like. Those skilled in the art can understand that the terminal device can include multiple baseband processors to adapt to different network modes, and the terminal device can include multiple central processors to enhance its processing capability, and various components of the terminal device can be connected by various buses. The baseband processor can also be referred to as a baseband processing circuit or a baseband processing chip. The central processor can also be referred to as a central processing circuit or a central processing chip. The function of processing communication protocols and communication data can be built into the processor, or stored in the memory in the form of a software program, and the processor executes the software program to realize the baseband processing function.

[0318] The memory is mainly used for storing software programs and data. The memory 812 can exist independently and be connected to the processor 811. Alternatively, the memory 812 can be integrated with the processor 811, for example, integrated in a chip. The memory 812 can store program codes for executing the technical solutions of the embodiments of the present application, and the processor 811 controls the execution. Various computer programs executed can also be regarded as a driver of the processor 811.

[0319] FIG. 8 only shows one memory and one processor. In actual terminal devices, there can be multiple processors and multiple memories. The memory can also be referred to as a storage medium or a storage device, etc. The memory can be a storage element on the same chip as the processor, that is, an on-chip storage element, or an independent storage element, and the embodiments of the present application do not limit this.

[0320] The transceiver 813 can be configured to support the receiving or transmitting of radio frequency signals between the communication device and a terminal. The transceiver 813 can be connected to the antenna 815. The transceiver 813 includes a transmitter Tx and a receiver Rx. Specifically, the one or more antennas 815 can receive radio frequency signals, the receiver Rx of the transceiver 813 is configured to receive the radio frequency signals from the antenna and 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 811 for further processing, such as demodulation processing and decoding processing, by the processor 811. In addition, the transmitter Tx in the transceiver 813 is also configured to receive modulated digital baseband signals or digital intermediate frequency signals from the processor 811, and convert the modulated digital baseband signals or digital intermediate frequency signals into radio frequency signals, and transmit the radio frequency signals through the one or more antennas 815. Specifically, the receiver Rx can selectively perform one or more levels of down-mixing processing and analog-to-digital conversion processing on the radio frequency signals to obtain the digital baseband signals or digital intermediate frequency signals, and the order of the down-mixing processing and the analog-to-digital conversion processing can be adjustable. The transmitter Tx can selectively perform one or more levels of up-mixing processing and digital-to-analog conversion processing on the modulated digital baseband signals or digital intermediate frequency signals to obtain the radio frequency signals, and the order of the up-mixing processing and the digital-to-analog conversion processing can be adjustable. The digital baseband signals and the digital intermediate frequency signals can be collectively referred to as digital signals.

[0321] The transceiver 813 can also be referred to as a transceiving unit, a transceiver, a transceiving device, etc. Optionally, the devices in the transceiving unit for implementing the receiving function can be regarded as a receiving unit, and the devices in the transceiving unit for implementing the transmitting function can be regarded as a transmitting unit, i.e., the transceiving unit includes the receiving unit and the transmitting unit, the receiving unit can also be referred to as a receiver, an input port, a receiving circuit, etc., and the transmitting unit can be referred to as a transmitter, a transmitter, or a transmitting circuit, etc.

[0322] It should be noted that the communication device 800 shown in FIG. 8 can be specifically configured to implement the steps implemented by the network device in the foregoing method embodiments, and achieve the corresponding technical effects of the network device. The specific implementation mode of the communication device 800 shown in FIG. 8 can be referred to the description in the foregoing method embodiments, which will not be described here one by one.

[0323] Please refer to FIG. 9, which is a structural schematic diagram of a communication device involved in the above embodiments provided by the embodiments of the present application.

[0324] It can be understood that the communication apparatus 900 includes, for example, modules, units, elements, circuits, or interfaces, and the like, which are appropriately configured together to perform the technical solutions provided in the present application. The communication apparatus 900 can be a terminal device or a network device described above, or can be a component (for example, a chip) of the devices, to implement the methods described in the following method embodiments. The communication apparatus 900 includes one or more processors 901. The processor 901 can be a general processor or a special-purpose processor, and the like. For example, it can be a baseband processor or a central processing unit. The baseband processor can be used to process communication protocols and communication data, and the central processing unit can be used to control the communication apparatus (such as a RAN node, a terminal, or a chip, and the like), execute software programs, and process data of the software programs.

[0325] Optionally, in one design, the processor 901 can include a program 903 (which can also be referred to as code or instructions at times) that can be run on the processor 901, so that the communication apparatus 900 performs the methods described in the following embodiments. In yet another possible design, the communication apparatus 900 includes a circuit (not shown in FIG. 9).

[0326] Optionally, the communication apparatus 900 can include one or more memories 902 having a program 904 (which can also be referred to as code or instructions at times) stored thereon, which can be run on the processor 901, so that the communication apparatus 900 performs the methods described in the above method embodiments.

[0327] Optionally, the processor 901 and / or the memory 902 can include an AI module 907, 908, which is used to implement AI-related functions. The AI module can be implemented in software, hardware, or a combination of software and hardware. For example, the AI module can include a radio intelligence control (RIC) module. For example, the AI module can be a near-real-time RIC or a non-real-time RIC.

[0328] Optionally, the processor 901 and / or the memory 902 can also store data. The processor and the memory can be separately arranged, or can be integrated together.

[0329] Optionally, the communication apparatus 900 can also include a transceiver 905 and / or an antenna 906. The processor 901 can also be referred to as a processing unit, which controls the communication apparatus (such as a RAN node or a terminal). The transceiver 905 can also be referred to as a transceiving unit, a transceiver, a transceiving circuit, or a transceiver, and the like, which is used to realize the transceiving function of the communication apparatus through the antenna 906.

[0330] The processing unit 501 in FIG. 5 can be the processor 901. The transceiving unit 502 in FIG. 5 can be a communication interface, which can be the transceiver 905 in FIG. 9, and the transceiver 905 can include an input interface and an output interface. Alternatively, the transceiver 905 can also be a transceiving circuit, which can include an input interface circuit and an output interface circuit.

[0331] The embodiments of the present application further provide a computer readable storage medium for storing one or more computer-executable instructions, which, when executed by a processor, cause the processor to perform the method described in the possible implementation manners of the first communication device or the second communication device.

[0332] The embodiments of the present application further provide a computer program product (or computer program), which, when executed by a processor, causes the processor to perform the method described in the possible implementation manners of the first communication device or the second communication device.

[0333] The embodiments of the present application further provide a chip system, which includes at least one processor for supporting the first communication device or the second communication device to perform the functions involved in the possible implementation manners of the communication device. Optionally, the chip system further includes an interface circuit for providing program instructions and / or data for the at least one processor. In a possible design, the chip system can further include a memory for storing necessary program instructions and data of the communication device. The chip system can be composed of a chip, or can include a chip and other discrete devices. The communication device can be the first communication device or the second communication device in the method embodiments.

[0334] The embodiments of the present application further provide a communication system, which includes the first communication device in any of the above embodiments.

[0335] Optionally, the communication system further includes the second communication device.

[0336] In several embodiments provided in the present application, it should be understood that the disclosed system, device and method can be implemented in other manners. For example, the described device embodiments are merely schematic. For example, the division of the units is only a logical function division. There can be another division manner for the actual implementation, for example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the displayed or discussed mutual couplings or direct couplings or communication connections between different units, can be indirect couplings or communication connections through some interfaces, devices or units, and can be in electrical, mechanical or other forms.

[0337] The units described as separated components can or can not be physically separated, and the components displayed as units can or can not be physical units, i.e., can be located in one place, or can be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purposes of the embodiments of the present application.

[0338] In addition, each functional unit in the embodiments of the present application can be integrated in one processing unit, or each unit can exist physically as a separate unit, or two or more units can be integrated in one unit. The integrated unit can be implemented in the form of hardware or in the form of a software functional unit. When the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application essentially or substantially, or all or part of the technical solutions, can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the embodiments of the present application. The foregoing storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), magnetic disk or optical disk, and various other media that can store program codes.

Claims

1. A communication method characterized by comprising: The method comprises: determining a first sub-model, the first sub-model comprising M model units, the M model units being part of N model units, M being a positive integer, N being greater than M; wherein the N model units are used to determine a first model, the first model comprising at least two sub-models, the first sub-model being one of the at least two sub-models; processing first input data based on the first sub-model to obtain first output data.

2. The method of claim 1, wherein, The method further comprises: sending or receiving first information, the first information being used to indicate the first sub-model or model information of the first sub-model.

3. The method according to claim 1 or 2, characterized in that, The method further comprises: receiving second information, the second information being used to indicate the first input data; wherein the first input data is used to determine the first sub-model.

4. The method according to any one of claims 1 to 3, characterized in that, The method further comprises: receiving third information, the third information being used to indicate the first model; or, obtaining the first model based on training data.

5. The method according to any one of claims 1 to 4, characterized in that, The method further comprises: receiving fourth information, the fourth information being used to indicate a second sub-model or model information of the second sub-model; The processing of the first input data based on the first sub-model to obtain the first output data comprises: in a case where state information of a first communication apparatus does not satisfy the second sub-model, processing the first input data based on the first sub-model to obtain the first output data.

6. The method of claim 5, wherein, The method further comprises: sending fifth information, the fifth information being used to indicate that the state information of the first communication apparatus does not satisfy the second sub-model.

7. The method according to any one of claims 1 to 6, characterized in that, The method further comprises: sending sixth information, the sixth information being used to indicate the first output data.

8. A communication method characterized by comprising: The method comprises: determining first information, the first information being used to indicate a first sub-model or model information of the first sub-model; wherein the first sub-model comprises M model units, the M model units being part of N model units, M being a positive integer, N being greater than M; the N model units being used to determine a first model, the first model comprising at least two sub-models, the first sub-model being one of the at least two sub-models; sending the first information.

9. A communication method characterized by comprising: The method comprises: receiving first information, the first information being used to indicate a first sub-model or model information of the first sub-model; wherein the first sub-model comprises M model units, the M model units being part of N model units, M being a positive integer, N being greater than M; the N model units being used to determine a first model, the first model comprising at least two sub-models, the first sub-model being one of the at least two sub-models; performing model management on the first sub-model based on the first information.

10. The method according to claim 8 or 9, characterized in that, The method further comprises: sending second information, the second information being used to indicate the first input data; wherein the first input data is used to determine the first sub-model.

11. The method according to any one of claims 8 to 10, characterized in that, The method further comprises: sending third information, the third information being used to indicate the first model; or, the first model being obtained by training of a first communication apparatus based on training data.

12. The method according to any one of claims 8 to 11, characterized in that, The method further comprises: transmit fourth information, the fourth information being used for indicating a second sub-model or model information of the second sub-model; receive fifth information, the fifth information being used for indicating that state information of the first communication apparatus does not satisfy the second sub-model.

13. The method according to any one of claims 8 to 12, characterized in that, The method further includes: receive sixth information, the sixth information being used for indicating the first output data.

14. The method according to any one of claims 1 to 13, characterized in that, The model information of any sub-model is pre-configured.

15. The method according to any one of claims 1 to 14, characterized in that, In the at least two sub-models, any sub-model contains one or more model units and a connection sequence, which are different from one or more model units and a connection sequence contained by other sub-models.

16. The method according to any one of claims 1 to 15, characterized in that, In the at least two sub-models, the model information of any sub-model includes at least one of the following: an index corresponding to the any sub-model, a use case corresponding to the any sub-model, an AI characteristic corresponding to the any sub-model, a function corresponding to the any sub-model, a task corresponding to the any sub-model, a data type of input data of the any sub-model, a data type of output data of the any sub-model, an identification of one or more model units contained by the any sub-model, a connection relationship between one or more model units contained by the any sub-model, or a target performance corresponding to the any sub-model.

17. The method according to any one of claims 1 to 16, characterized in that, In the at least two sub-models, target performances corresponding to different sub-models are different.

18. A communications device, characterized by include a module for performing the method according to any one of claims 1 to 17.

19. A communications device, characterized by include at least one processor for performing the method according to any one of claims 1 to 17.

20. The communication apparatus according to claim 19, wherein, The communication apparatus is a chip or a chip system.

21. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a computer program or instructions, when the computer program or instructions are executed by a communication apparatus, a method according to any one of claims 1 to 17 is implemented.

22. A computer program product, characterised in that, include a computer program or instructions, when the computer program or instructions are executed by a computer, a method according to any one of claims 1 to 17 is implemented.

Images