Communication method and related equipment

Abstract

The invention discloses a communication method and related equipment, which are used for enabling the computing power of a communication node to be used for the acquisition of artificial intelligence (AI) data. In the method, after a first node receives first configuration information, the first node can perform AI data acquisition based on the first configuration information to obtain first AI data; thereafter, the first node can send the first AI data, and the subsequent receiver of the first AI data can perform model processing of the first AI model based on the first AI data. Therefore, under the condition that the communication node in the communication system is used as the AI participation node, the computing power of the communication node can be used as the acquisition node of the AI data, and the acquisition of the AI data is realized.

Description

Technical Field

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

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

[0003] Currently, in a wireless communication system, communication nodes generally have signal transceiver capabilities and computing capabilities. Taking a network device with computing capabilities as an example, the computing capabilities of the network device mainly provide computing power support for the signal transceiver capabilities (for example, performing transmission processing and reception processing on signals) to achieve communication between the network device and other communication nodes.

[0004] However, in addition to processing communication signals in the communication network, communication nodes may also need to take into account artificial intelligence (AI) related processing.

[0005] Therefore, how to achieve the integration of AI related processing and the communication network is a technical problem to be solved urgently. Summary of the Invention

[0006] This application provides a communication method and related devices for enabling a communication node to act as a collection node for AI data and realizing the collection of AI data.

[0007] In a first aspect of this application, a communication method is provided. This method is executed by a first node. The first node can be a communication device (such as a network device or a terminal device), or the first node can be a part of the components in the communication device (such as a processor, a chip, or a chip system, etc.), or the first node can also be a logical module or software that can implement all or part of the functions of the communication device. In this method, the first node receives first configuration information, and the first configuration information is used to configure AI data collection; the first node sends first AI data, and the first AI data is collected based on the first configuration information; wherein, the first AI data is used for model processing of a first AI model.

[0008] Based on the above technical solution, after receiving the first configuration information, the first node can collect AI data based on the first configuration information to obtain first AI data; thereafter, the first node can send the first AI data, and the subsequent recipient of the first AI data can perform model processing of the first AI model based on the first AI data. Thus, when the communication node in the communication system acts as an AI participating node, the communication node can act as a collection node for AI data and realize the collection of AI data.

[0009] In addition, as a communication node, after the first node sends the first AI data, the recipient of the first AI data can perform model processing of the AI model based on the AI data collected by the communication node.

[0010] It should be noted that the technical solution provided in this application can be applied to a communication system, which may include N distributed nodes and a control node, where N is a positive integer. Among them, any one of the N distributed nodes can be used as the first node to execute the methods in the first aspect and its possible implementation manners; the control node can be used to execute the methods in the second aspect and its possible implementation manners described later.

[0011] Optionally, the communication system may further include a central node, and the central node can be used to execute the methods in the third aspect and its possible implementation manners described later. Alternatively, the function of the central node is executed by the control node, that is, the control node is further used to execute the methods related to the central node.

[0012] Optionally, the communication system may further include a data receiving node, and the data receiving node can be used to execute the methods in the fourth aspect and its possible implementation manners described later. Alternatively, the function of the data receiving node is executed by the control node, that is, the control node is further used to execute the methods related to the data receiving node.

[0013] In this application, terms such as AI model, neural network model, AI neural network model, machine learning model, and AI processing model can be replaced with each other.

[0014] In this application, terms such as data acquisition, data collection, data gathering, data obtaining, and data capture can be replaced with each other.

[0015] It should be understood that the model processing involved in the embodiments of this application includes at least one of model training, model inference, and model monitoring. Correspondingly, the first AI data sent by the first node includes at least one of the AI data used in the model training stage of the first AI model, the AI data used in the model inference stage of the first AI model, and the AI data used in the model monitoring stage of the first AI model.

[0016] In one implementation example, the AI data used in the model training stage of the first AI model included in the first AI data may include at least one of the input data, feature data, and label data for training the first AI model.

[0017] In another implementation example, the AI data used in the model inference stage of the first AI model included in the first AI data may include at least one of input data, feature data, and inference result data for the inference of the first AI model.

[0018] In another implementation example, the AI data used in the model monitoring stage of the first AI model included in the first AI data may include at least one of input data, feature data, label data, inference result data, AI model performance data, and communication performance data for the monitoring of the first AI model.

[0019] Optionally, different communication nodes (such as the first node and other nodes mentioned later, including the second node, control node, central node, etc.) can transmit wireless communication signals (such as the transceiver of configuration information of communication resources, the transceiver of reference signals, etc.). The AI models involved in the embodiments of the present application (such as the first AI model, the second AI model mentioned later, etc.) can be used to process the wireless communication signals (including at least one of management, configuration, update, and optimization). For example, the AI model may include an AI model for modulation and / or demodulation, an AI model for channel prediction, an AI model for beam management, an AI model for assisted positioning, an AI model for channel compression, an AI model for resource scheduling, an AI model for mobility management, an AI model for load balancing, an AI model for network energy saving, and an AI model for replacing one or more modules in a transmitter and / or receiver. Or, the AI models involved in the embodiments of the present application may also be AI models for other AI tasks, such as an AI model for image recognition, an AI model for natural language processing, an AI model for computer vision, etc.

[0020] In a possible implementation manner of the first aspect, the first configuration information includes at least one of the following:

[0021] The identifier of the AI model corresponding to the collected AI data;

[0022] The identifier of the AI function corresponding to the collected AI data;

[0023] The identifier of the source node of the collected AI data;

[0024] The indication information indicating the model processing corresponding to the collected AI data;

[0025] The indication information indicating that the configuration method of the configuration information for AI data collection is centralized or decentralized;

[0026] The indication information indicating that the AI data collection method is centralized or decentralized;

[0027] Indication information indicating the AI data characteristics of the collected AI data;

[0028] Indication information indicating the AI data processing of the collected AI data;

[0029] Indication information indicating the collection period of the AI data;

[0030] Indication information indicating the transmission information of the collected AI data.

[0031] Optionally, the AI data characteristics include one or more of the quantity of the AI data, the sample size, the collection time, the collection location, and the distribution.

[0032] Optionally, the AI data processing includes post-processing of the output data of the AI model and / or pre-processing of the input data of the AI model, such as one or more of dimension conversion and precision conversion.

[0033] Optionally, the collection period of the AI data includes that the AI data is collected in a periodic manner, the AI data is collected in a semi-static manner, or the AI data is collected in a (dynamic) trigger manner, etc.

[0034] Optionally, the transmission information of the AI data includes one or more of the data structure, format, precision, dimension, and transmission resources.

[0035] It should be understood that when the first configuration information includes at least one of the above, the first configuration information can be sent through one or more messages, that is, the first node can obtain the first configuration information through the receiving process of one or more messages.

[0036] Based on the above technical solution, the first configuration information for configuring AI data collection can include at least one of the above to improve the flexibility of the solution implementation.

[0037] In a possible implementation manner of the first aspect, the first node is one of the N distributed nodes, and N is an integer greater than or equal to 1.

[0038] Based on the above technical solution, the first node can be one of the N distributed nodes, that is, any one of the N distributed nodes can execute the method executed by the first node, so that the N distributed nodes can all be used as data collection nodes to implement AI data collection in a distributed scenario.

[0039] In a possible implementation of the first aspect, the first node receives first configuration information, including: the first node receives the first configuration information from a control node, where the control node is used to control data collection of the N distributed nodes; or, the first node receives the first configuration information from the control node through a central node; or, the first node receives the first configuration information from a second node, where the second node is a node different from the first node among the N distributed nodes, and N is greater than 1.

[0040] Optionally, before the second node sends the first configuration information to the first node, the second node may receive one or more configuration information from the control node and send the first configuration information in the one or more configuration information to the first node. In other words, the first configuration information sent by the second node to the first node is from the control node.

[0041] Optionally, the control node and the central node may be the same node, or the control node and the central node may be different logical nodes in physical nodes, or the control node and the central node may be two independent different nodes.

[0042] Based on the above technical solution, the first node can receive the first configuration information through the above multiple methods. The first node is one of the N distributed nodes. In this way, the distributed nodes can receive the first configuration information in multiple different scenarios, and the flexibility of the solution implementation is improved.

[0043] In a possible implementation of the first aspect, the first configuration information is the configuration information corresponding to the first node among the M configuration information. The M configuration information is at least used to configure AI data collection of M distributed nodes among the N distributed nodes, and M is less than or equal to N; the method further includes: the first node sends at least one of the M configuration information to at least one of the M distributed nodes.

[0044] Based on the above technical solution, the first node can receive the M configuration information and determine the first configuration information among the M configuration information. Among them, the M configuration information is at least used to configure AI data collection of M distributed nodes among the N distributed nodes. Correspondingly, the first node can send at least one of the M configuration information to other nodes among the M distributed nodes, so that other distributed nodes can obtain the corresponding configuration information and perform AI data collection.

[0045] Optionally, the first node sends at least one of the M configuration messages to at least one of the M distributed nodes, so that each of the M distributed nodes can obtain its corresponding configuration message, facilitating the M distributed nodes to implement AI data collection based on their corresponding configuration messages. Further optionally, during the sending process, the first node may send the configuration message corresponding to each of the M distributed nodes to each of the M distributed nodes, or the first node may send the M configuration messages to each of the M distributed nodes, or the first node may send the M configuration messages to some of the M distributed nodes, and the some distributed nodes send the configuration messages corresponding to the other some distributed nodes to the other some distributed nodes, or the M distributed nodes can obtain their corresponding configuration messages through other means, which is not limited here.

[0046] Optionally, among the N distributed nodes, the configuration messages corresponding to different distributed nodes may be different. Therefore, the M configuration messages can be respectively used to configure the AI data collection of M distributed nodes among the N distributed nodes.

[0047] Optionally, among the N distributed nodes, the configuration messages corresponding to different distributed nodes may be the same. Therefore, in addition to the M configuration messages being respectively used to configure the AI data collection of M distributed nodes among the N distributed nodes, at least one of the M configuration messages can also be used to configure the AI data collection of at least one other distributed node among the N distributed nodes except the M distributed nodes.

[0048] In a possible implementation manner of the first aspect, the first node is one of the N distributed nodes, where N is an integer greater than or equal to 1; the first node sending the first AI data includes: the first node sending the first AI data to a data receiving node for AI data collection; or, the first node sending the first AI data to a central node.

[0049] Optionally, the control node and the data receiving node may be the same node, or the control node and the data receiving node may be different logical nodes in a physical node, or the control node and the data receiving node may be two independent different nodes.

[0050] Based on the above technical solution, the first node can send the first AI data through the above multiple methods. The first node is one of the N distributed nodes. In this way, the distributed nodes can send the collected AI data in multiple different scenarios, improving the flexibility of the solution implementation.

[0051] In a possible implementation of the first aspect, the method further includes: the first node sending the data of the first node to other nodes among the N distributed nodes, where the data of the first node is used for data collection by the other nodes.

[0052] Optionally, the data of the first node may include part or all of the first AI data.

[0053] Optionally, the data of the first node may include the communication data of the first node, such as reference signals, positioning data, etc.

[0054] Optionally, the data of the first node may include the model data of the local AI model of the first node, such as at least one of the AI data used in the model training stage of the local AI model, the AI data used in the model inference stage of the local AI model, and the AI data used in the model monitoring stage of the local AI model.

[0055] Based on the above technical solution, for the N distributed nodes, part or all of the AI data collected by one of the distributed nodes can be determined by the data sent by other distributed nodes. Correspondingly, the first node may also send the data of the first node to other nodes among the N distributed nodes, so that the other nodes can implement data collection based on the data of the first node.

[0056] In a possible implementation of the first aspect, the method further includes: the first node receiving the data from other nodes among the N distributed nodes, where the data of the other nodes is used to determine part or all of the first AI data. Based on the above technical solution, for the N distributed nodes, part or all of the AI data collected by one of the distributed nodes can be determined by the data sent by other distributed nodes. Correspondingly, the first node may also receive the data from other nodes among the N distributed nodes (such as the second data from the second node), so that the first node can implement data collection based on the data of the other nodes.

[0057] Optionally, the data of the other nodes may include the communication data of the other nodes, such as reference signals, positioning data, etc.

[0058] Optionally, the data of the other node may include model data of the local AI model of the other node, such as at least one of the AI data used in the model training phase of the local AI model, the AI data used in the model inference phase of the local AI model, and the AI data used in the model monitoring phase of the local AI model. Exemplarily, taking the other node as the second node, the local AI model of the second node may be denoted as the second AI model, and the model data of the second AI model may be denoted as the second AI data. Among them, the implementation of the second AI data is similar to that of the first AI data. The second AI data received by the first node may include at least one of the AI data used in the model training phase of the second AI model, the AI data used in the model inference phase of the second AI model, and the AI data used in the model monitoring phase of the second AI model.

[0059] In one implementation example, the AI data used in the model training phase of the second AI model included in the second AI data includes at least one of the input data, feature data, and label data used for the training of the second AI model;

[0060] In another implementation example, the AI data used in the model inference phase of the second AI model included in the second AI data includes at least one of the input data, feature data, and inference result data used for the inference of the second AI model;

[0061] In another implementation example, the AI data used in the model monitoring phase of the second AI model included in the second AI data includes at least one of the input data, feature data, label data, inference result data, AI model performance data, and communication performance data used for the monitoring of the second AI model.

[0062] In the second aspect of the present application, a communication method is provided. This method is executed by a control node. The control node may be a communication device (such as a network device or a terminal device), or the control node may be a part of the components in the communication device (such as a processor, a chip, or a chip system, etc.), or the control node may also be a logical module or software that can implement all or part of the communication device functions. In this method, the control node determines first configuration information for AI data collection, and the control node sends the first configuration information.

[0063] Based on the above technical solution, the first configuration information sent by the control node is used for AI data collection. That is, after the first node receives the first configuration information, the first node can perform AI data collection based on the first configuration information to obtain the first AI data. Thereafter, the first node can send the first AI data, and the subsequent recipient of the first AI data can perform model processing of the first AI model based on the first AI data. Thus, when the communication nodes in the communication system act as AI participating nodes, the communication nodes can act as AI data collection nodes to achieve the collection of AI data.

[0064] In a possible implementation manner of the second aspect, the first configuration information includes at least one of the following:

[0065] The identifier of the AI model corresponding to the collected AI data;

[0066] The identifier of the AI function corresponding to the collected AI data;

[0067] The identifier of the source node of the collected AI data;

[0068] The indication information indicating the model processing corresponding to the collected AI data;

[0069] The indication information indicating that the configuration method of the configuration information for AI data collection is centralized or decentralized;

[0070] The indication information indicating that the AI data collection method is centralized or decentralized;

[0071] The indication information indicating the AI data characteristics of the collected AI data;

[0072] The indication information indicating the AI data processing of the collected AI data;

[0073] The indication information indicating the collection period of the AI data;

[0074] The indication information indicating the transmission information of the collected AI data.

[0075] Based on the above technical solution, the first configuration information for configuring AI data collection may include at least one of the above to improve the flexibility of the solution implementation.

[0076] Optionally, the AI data characteristics include one or more of the quantity, sample size, collection time, collection location, and distribution of the AI data.

[0077] Optionally, the AI data processing includes post-processing of the output data of the AI model and / or pre-processing of the input data of the AI model, such as one or more of dimension conversion and precision conversion.

[0078] Optionally, the acquisition period of AI data includes acquiring AI data in a periodic manner, acquiring AI data in a semi-static manner, or acquiring AI data in a (dynamic) trigger manner, etc.

[0079] Optionally, the transmission information of AI data includes one or more of data structure, format, precision, dimension, and transmission resources.

[0080] It should be understood that when the first configuration information includes at least one of the above, the first configuration information can be sent through one or more messages, that is, the first node can obtain the first configuration information through the receiving process of one or more messages.

[0081] In a possible implementation manner of the second aspect, the control node is used to control the data acquisition of N distributed nodes, where N is an integer greater than or equal to 1; the first configuration information is used for the AI data acquisition of the first AI node among the N distributed nodes.

[0082] Based on the above technical solution, the first node that performs data acquisition based on the first configuration information can be one of the N distributed nodes, that is, any one of the N distributed nodes can execute the method executed by the first node, so that the N distributed nodes can all be used as data acquisition nodes to achieve AI data acquisition in a distributed scenario.

[0083] In a possible implementation manner of the second aspect, the control node sending the first configuration information includes: the control node sending the first configuration information to the first node; or, the control node sending the first configuration information to the first node through the central node; or, the control node sending the first configuration information to the first node through the second node, where the second node is a node different from the first node among the N distributed nodes and N is greater than 1.

[0084] Based on the above technical solution, the control node can send the first configuration information to the first node through the above multiple methods. The first node is one of the N distributed nodes. In this way, the distributed nodes can receive the first configuration information in multiple different scenarios and improve the flexibility of the scheme implementation.

[0085] In a possible implementation manner of the second aspect, the first configuration information is one of the K configuration information sent, and the K configuration information are respectively used to configure the AI data acquisition of N distributed nodes, where K is less than or equal to N.

[0086] Based on the above technical solution, the control node can send K configuration information, where the K configuration information is respectively used to configure AI data collection in N distributed nodes. Correspondingly, each distributed node can obtain corresponding configuration information based on the K configuration information and perform AI data collection.

[0087] Optionally, among the N distributed nodes, the configuration information corresponding to different distributed nodes may be different from each other. For this reason, the values of K and N can be equal, that is, the K configuration information can be respectively used to configure AI data collection of different nodes in the N distributed nodes.

[0088] Optionally, among the N distributed nodes, the configuration information corresponding to different distributed nodes may be the same. For this reason, K can be less than N, that is, at least one of the K configuration information is used to configure AI data collection of at least two nodes in the N distributed nodes. Correspondingly, the configuration information of the AI data collection of the at least two nodes is the same.

[0089] In a possible implementation manner of the second aspect, the method further includes: the control node receives first AI data, and the first AI data is collected based on the first configuration information.

[0090] Based on the above technical solution, after the control node sends the first configuration information, the control node can also receive the first AI data collected based on the first configuration information. Among them, the first AI data can come from the first node. The first node is used as a communication node. After the first node sends the AI data, the control node can implement model processing of the AI model based on the AI data collected by the communication node.

[0091] It should be understood that the model processing involved in the embodiments of the present application includes at least one of model training, model inference, and model monitoring. Correspondingly, the first AI data sent by the first node includes at least one of the AI data used in the model training stage of the first AI model, the AI data used in the model inference stage of the first AI model, and the AI data used in the model monitoring stage of the first AI model.

[0092] In one implementation example, the AI data used in the model training stage of the first AI model included in the first AI data may include at least one of input data, feature data, and label data for training the first AI model.

[0093] In another implementation example, the AI data used in the model inference stage of the first AI model included in the first AI data may include at least one of input data, feature data, and inference result data for inferring the first AI model.

[0094] In another implementation example, the AI data used in the model monitoring stage of the first AI model included in the first AI data may include at least one of input data, feature data, label data, inference result data, AI model performance data, and communication performance data for monitoring the first AI model.

[0095] In a possible implementation manner of the second aspect, the control node receives the first AI data, including: the control node receives the first AI data from the first node; or, the control node receives the first AI data through the central node.

[0096] Based on the above technical solutions, the control node can receive the first AI data through the above various methods to improve the flexibility of the solution implementation.

[0097] The third aspect of this application provides a communication method, which is executed by the central node. The central node may be a communication device (such as a network device or a terminal device), or the central node may be a part of the components in the communication device (such as a processor, a chip, or a chip system, etc.), or the central node may also be a logical module or software that can implement all or part of the communication device functions. In this method, the central node receives the first configuration information from the control node; the central node sends the first configuration information to the first node.

[0098] Among them, the steps executed by the central node may also refer to the descriptions in the first aspect or the second aspect and their possible implementation manners above.

[0099] The fourth aspect of this application provides a communication method, which is executed by the data receiving node. The data receiving node may be a communication device (such as a network device or a terminal device), or the data receiving node may be a part of the components in the communication device (such as a processor, a chip, or a chip system, etc.), or the data receiving node may also be a logical module or software that can implement all or part of the communication device functions. In this method, the data receiving node receives the first AI data from the first node.

[0100] Among them, the steps executed by the data receiving node may also refer to the descriptions in the first aspect or the second aspect and their possible implementation manners above.

[0101] The fifth aspect of this application provides a communication device, which is the first node or a part of the components in the first node (such as a processor, a chip, a chip system, a logical module, or software, etc.). The device includes a transceiver unit and a processing unit; the transceiver unit is used to receive the first configuration information, and the first configuration information is used to configure AI data acquisition; the processing unit sends the first AI data, and the first AI data is collected based on the first configuration information; among them, the first AI data is used for model processing of the first AI model.

[0102] In the fifth aspect of the present application, the constituent modules of the communication device can also be used to execute the steps performed in each possible implementation manner of the first aspect, and achieve the corresponding technical effects. For details, reference can be made to the first aspect, which will not be elaborated here.

[0103] In the sixth aspect of the present application, a communication device is provided. The device is a control node or a part of the components in the control node (such as a processor, a chip, a chip system, a logic module, or software, etc.). The device includes a transceiver unit and a processing unit. The processing unit is used to determine first configuration information for AI data collection, and the transceiver unit is used to send the first configuration information.

[0104] In the sixth aspect of the present application, the constituent modules of the communication device can also be used to execute the steps performed in each possible implementation manner of the second aspect, and achieve the corresponding technical effects. For details, reference can be made to the second aspect, which will not be elaborated here.

[0105] In the seventh aspect of the present application, a communication device is provided. The device is a central node or a part of the components in the central node (such as a processor, a chip, a chip system, a logic module, or software, etc.). The device includes a transceiver unit. The transceiver unit is used to receive the first configuration information from the control node, and the transceiver unit is also used to send the first configuration information to the first node.

[0106] In the seventh aspect of the present application, the constituent modules of the communication device can also be used to execute the steps performed in each possible implementation manner of the third aspect, and achieve the corresponding technical effects. For details, reference can be made to the third aspect, which will not be elaborated here.

[0107] In the eighth aspect of the present application, a communication device is provided. The device is a data receiving node or a part of the components in the data receiving node (such as a processor, a chip, a chip system, a logic module, or software, etc.). The device includes a transceiver unit, and the transceiver unit is used to receive the first AI data from the first node.

[0108] In the eighth aspect of the present application, the constituent modules of the communication device can also be used to execute the steps performed in each possible implementation manner of the fourth aspect, and achieve the corresponding technical effects. For details, reference can be made to the fourth aspect, which will not be elaborated here.

[0109] In the ninth aspect of the present application, a communication device is provided, including at least one processor. The at least one processor is coupled to a memory. The memory is used to store programs or instructions. The at least one processor is used to execute the programs or instructions so that the device implements the method described in any one of the possible implementation manners of any one of the first aspect to the fourth aspect.

[0110] The tenth aspect of the present application provides a communication device, including at least one logic circuit and an input / output interface; the logic circuit is configured to execute the method described in any one of the possible implementation manners in any one of the foregoing first aspect to fourth aspect.

[0111] The eleventh aspect of the present application provides a communication system, which includes the above-mentioned first node and a control node.

[0112] Optionally, the communication system further includes other nodes among the N distributed nodes, such as the second node.

[0113] Optionally, the communication system further includes a data receiving node.

[0114] Optionally, the communication system further includes a central node.

[0115] The twelfth aspect of the present application provides a computer-readable storage medium, which is used to store one or more computer-executable instructions. When the computer-executable instructions are executed by a processor, the processor executes the method described in any one of the possible implementation manners in any one of the foregoing first aspect to fourth aspect.

[0116] The thirteenth 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 described in any one of the possible implementation manners in any one of the foregoing first aspect to fourth aspect.

[0117] The fourteenth aspect of the present application provides a chip system, which includes at least one processor for supporting a communication device to implement the method described in any one of the possible implementation manners in any one of the foregoing first aspect to fourth aspect.

[0118] In a possible design, the chip system may further include a memory for storing necessary program instructions and data of the communication device. The chip system may be composed of chips or may include chips and other discrete devices. Optionally, the chip system further includes an interface circuit, and the interface circuit provides program instructions and / or data for the at least one processor.

[0119] Among them, the technical effects brought by any one of the design manners in the fifth aspect to the fourteenth aspect can be referred to the technical effects brought by different design manners in the foregoing first aspect to the fourth aspect, and will not be elaborated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0120] Figures 1a to 1c It is a schematic diagram of the communication system provided by the present application;

[0121] Figures 2a to 2hSchematic diagram of the AI processing process involved in this application;

[0122] Figure 3 An interaction schematic diagram of the communication method provided by this application;

[0123] Figures 4a to 4e 、 Figures 5a to 5c 、 Figure 6 Schematic diagram of the AI processing process provided by this application;

[0124] Figures 7 to 11 Schematic diagram of the communication device provided by this application. Detailed implementation manners

[0125] First, some terms in the embodiments of this application are explained to facilitate the understanding of those skilled in the art.

[0126] (1) Terminal device: It can be a wireless terminal device capable of receiving scheduling and indication information from a network device. The wireless terminal device can be a device that provides voice and / or data connectivity to a user, or a handheld device with a wireless connection function, or other processing devices connected to a wireless modem.

[0127] A terminal device can communicate with one or more core networks or the Internet via a radio access network (RAN). The terminal device can be a mobile terminal device, such as a mobile phone (or a "cellular" phone, a mobile phone), a computer, and a data card. For example, it can be a portable, pocket-sized, handheld, computer-integrated, or vehicle-mounted mobile device that exchanges voice and / or data with the radio access network. For example, devices such as personal communication service (PCS) phones, cordless phones, session initiation protocol (SIP) phones, wireless local loop (WLL) stations, personal digital assistants (PDAs), tablets (Pads), computers with wireless transceiver functions, etc. A wireless terminal device can also be referred to as a system, a subscriber unit, a subscriber station, a mobile station, a mobile station (MS), a remote station, an access point (AP), a remote terminal device, an access terminal device, a user terminal device, a user agent, a subscriber station (SS), a customer premises equipment (CPE), a terminal, a user equipment (UE), a mobile terminal (MT), etc.

[0128] By way of example and not limitation, in the embodiments of the present application, the terminal device may also be a wearable device. Wearable devices, also known as wearable intelligent devices or smart wearable devices, etc., are the general term for devices developed by applying wearable technology to the intelligent design of daily wear, such as glasses, gloves, watches, clothing, and shoes. Wearable devices are portable devices that are directly worn on the body or integrated into the user's clothes or accessories. Wearable devices are not only a kind of hardware device, but also achieve powerful functions through software support, data interaction, and cloud interaction. Broadly speaking, wearable intelligent devices include those with complete functions and large sizes that can achieve complete or partial functions without relying on a smart phone, such as smart watches or smart glasses, etc., and those that only focus on a certain type of application function and need to cooperate with other devices such as smart phones, such as various smart bracelets for physical sign monitoring, smart helmets, and smart jewelry.

[0129] The terminal may also be a drone, a robot, a terminal in device-to-device (D2D) communication, a terminal in vehicle-to-everything (V2X) communication, a virtual reality (VR) terminal device, an augmented reality (AR) terminal device, a wireless terminal in industrial control, a wireless terminal in self-driving, a wireless terminal in remote medical, a wireless terminal in smart grid, a wireless terminal in transportation safety, a wireless terminal in smart city, a wireless terminal in smart home, etc.

[0130] In addition, the terminal device may also be a terminal device in a communication system evolved after the fifth-generation (5G) communication system (such as the sixth-generation (6G) communication system, etc.) or a terminal device in a future evolved public land mobile network (PLMN). Exemplarily, the 6G network can further expand the form and function of 5G communication terminals. 6G terminals include, but are not limited to, vehicles, cellular network terminals (integrating satellite terminal functions), drones, and Internet of Things (IoT) devices.

[0131] In the embodiments of the present application, the above terminal device may also obtain AI services provided by a network device. Optionally, the terminal device may also have AI processing capabilities.

[0132] (2) Network device: It can be a device in a wireless network. For example, the network device can be a RAN node (or device) that connects a terminal device to a wireless network, and can also be called a base station. Currently, some examples of RAN devices are: base station, evolved NodeB (eNodeB), gNB (gNodeB) in a 5G communication system, transmission reception point (TRP), evolved Node B (eNB), radio network controller (RNC), Node B (NB), home base station (e.g., home evolved Node B, or home Node B, HNB), base band unit (BBU), or wireless fidelity (Wi-Fi) access point AP, etc. Additionally, in a network structure, the network device can include a centralized unit (CU) node, or a distributed unit (DU) node, or a RAN device including a CU node and a DU node.

[0133] Optionally, the RAN node can also be a macro base station, a micro base station or an indoor station, a relay node or a donor node, or a radio controller in a cloud radio access network (CRAN) scenario. The RAN node can also be a server, a wearable device, a vehicle or an in-vehicle device, etc. For example, the access network device in vehicle to everything (V2X) technology can be a road side unit (RSU).

[0134] In another possible scenario, multiple RAN nodes cooperate to assist a terminal in achieving wireless access, and different RAN nodes respectively implement some functions of a base station. For example, the RAN node can be a central unit (CU), a distributed unit (DU), a CU-control plane (CP), a CU-user plane (UP), or a radio unit (RU), etc. The CU and the DU can be set separately, or can also be included in the same network element, such as a baseband unit (BBU). The RU can be included in a radio frequency device or a radio frequency unit, such as included in a remote radio unit (RRU), an active antenna unit (AAU), or a remote radio head (RRH).

[0135] In different systems, the CU (or CU-CP and CU-UP), DU, or RU may also have different names, but those skilled in the art can understand their meanings. For example, in an open RAN (O-RAN or ORAN) system, the CU can also be called an O-CU (open CU), the DU can also be called an O-DU, the CU-CP can also be called an O-CU-CP, the CU-UP can also be called an O-CU-UP, and the RU can also be called an O-RU. For the convenience of description, in this application, the CU, CU-CP, CU-UP, DU, and RU are used as examples for description. Any one of the CU (or CU-CP, CU-UP), DU, and RU in this application can be implemented by a software module, a hardware module, or a combination of a software module and a hardware module.

[0136] The communication between the access network device and the terminal device follows a certain protocol layer structure. The protocol layer may include a control plane protocol layer and a user plane protocol layer. The control plane protocol layer may include at least one of the following: Radio Resource Control (RRC) layer, Packet Data Convergence Protocol (PDCP) layer, Radio Link Control (RLC) layer, Media Access Control (MAC) layer, or Physical (PHY) layer, etc. The user plane protocol layer may include at least one of the following: Service Data Adaptation Protocol (SDAP) layer, PDCP layer, RLC layer, MAC layer, or Physical layer, etc.

[0137] For the correspondence between the network elements in the ORAN system and their achievable protocol layer functions, reference can be made to Table 1 below.

[0138] Table 1

[0139] ORAN network element Protocol layer functions of 3GPP O-CU-CP RRC+PCDP - Control Plane (PDCP-C) O-CU-UP SDAP+PCDP - User Plane (PDCP-U) O-DU RLC+MAC+PHY-high O-RU PHY-low

[0140] The network device may be other devices that provide wireless communication functions for the terminal device. The specific technologies and specific device forms adopted by the network device are not limited in the embodiments of the present application. For ease of description, the embodiments of the present application do not limit.

[0141] The network device may further include a core network device. For example, the core network device includes a mobility management entity (MME), a home subscriber server (HSS), a serving gateway (S-GW), a policy and charging rules function (PCRF), and a public data network gateway (PDN gateway, P-GW) in a 4th generation (4G) network; network elements such as an access and mobility management function (AMF), a user plane function (UPF), or a session management function (SMF) in a 5G network. In addition, the core network device may further include other core network devices in a 5G network and the next-generation network of the 5G network.

[0142] In the embodiments of the present application, the above network device may further have a network node with AI capabilities, which can provide AI services for terminals or other network devices. For example, it can be an AI node, a computing power node, a RAN node with AI capabilities, or a core network element with AI capabilities on the network side (access network or core network).

[0143] In the embodiments of the present application, the device for implementing the functions of the network device may be the network device or a device capable of supporting the network device to implement such functions, such as a chip system. This device may be installed in the network device. In the technical solutions provided in the embodiments of the present application, the device for implementing the functions of the network device is taken as an example of the network device to describe the technical solutions provided in the embodiments of the present application.

[0144] (3) Configuration and pre-configuration: In the present application, both configuration and pre-configuration are used. Among them, configuration means that the network device / server sends the configuration information or value of some parameters to the terminal through a message or signaling, so that the terminal can determine the communication parameters or resources during transmission according to these values or information. Pre-configuration is similar to configuration. It can be parameter information or parameter values pre-negotiated between the network device / server and the terminal device, or parameter information or parameter values specified by a standard protocol for the base station / network device or terminal device, or parameter information or parameter values pre-stored in the base station / server or terminal device. The present application does not limit this.

[0145] Furthermore, these values and parameters can be changed or updated.

[0146] (4) The terms "system" and "network" in the embodiments of the present application can be used interchangeably. "Plurality" means two or more. "And / or" describes the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, or B exists alone, where A and B can be singular or plural. The character " / " generally represents an "or" relationship between the associated objects before and after. "At least one (item)" or its similar expression refers to any combination of these items, including any combination of a single item or plural items. For example, "at least one of A, B, and C" includes A, B, C, AB, AC, BC, or ABC. Also, unless otherwise specified, the ordinal numbers such as "first" and "second" mentioned in the embodiments of the present application are used to distinguish multiple objects and are not used to limit the order, time sequence, priority, or importance of multiple objects.

[0147] (5) "Sending" and "receiving" in the embodiments of the present application represent the direction of signal transmission. For example, "sending information to XX" can be understood as the destination of the information being XX, which can include directly sending through the air interface or indirectly sending through other units or modules via the air interface. "Receiving information from YY" can be understood as the source of the information being YY, which can include directly receiving from YY through the air interface or indirectly receiving from YY through other units or modules via the air interface. "Sending" can also be understood as the "output" of the chip interface, and "receiving" can also be understood as the "input" of the chip interface.

[0148] In other words, sending and receiving can be carried out between devices. For example, between a network device and a terminal device, or can be carried out within a device. For example, sending or receiving between components, modules, chips, software modules, or hardware modules within a device through a bus, trace, or interface.

[0149] It can be understood that necessary processing may be performed on the information between the source and destination of the information transmission, such as encoding, modulation, etc., but the destination can understand the valid information from the source. Similar expressions in the present application can be understood similarly and will not be elaborated further.

[0150] (6) In the embodiments of the present application, "indication" may include direct indication and indirect indication, and may also include explicit indication and implicit indication. If the information indicated by a certain piece of information (such as the indication information described below) is called the information to be indicated, then in the specific implementation process, there are many ways to indicate the information to be indicated. For example, but not limited to, the information to be indicated can be directly indicated, such as the information to be indicated itself or the index of the information to be indicated, etc. It is also possible to indirectly indicate the information to be indicated by indicating other information, where there is an association relationship between the other information and the information to be indicated; it is also possible to only indicate a part of the information to be indicated, while the other parts of the information to be indicated are known or pre-agreed. For example, the arrangement order of each piece of information pre-agreed (such as protocol pre-definition) can be used to implement the indication of specific information, thereby reducing the indication overhead to a certain extent. The present application does not limit the specific manner of indication. It can be understood that for the sender of the indication information, the indication information can be used to indicate the information to be indicated, and for the receiver of the indication information, the indication information can be used to determine the information to be indicated.

[0151] In the present application, unless otherwise specified, the same or similar parts between various embodiments can be referred to each other. In various embodiments of the present application, as well as in each method / design / implementation manner in each embodiment, if there is no special description and logical conflict, the terms and / or descriptions between different embodiments, as well as between each method / design / implementation manner in each embodiment, are consistent and can be referenced to each other. The technical features in different embodiments, as well as in each method / design / implementation manner in each embodiment, can be combined to form new embodiments, methods, or implementation manners according to their internal logical relationships. The embodiments of the present application described below do not constitute a limitation on the protection scope of the present application.

[0152] 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.). Among them, the communication system includes at least one network device and / or at least one terminal device.

[0153] Please refer to Figure 1a , which is a schematic diagram of the communication system in the present application. Figure 1a shows, by way of example, one network device and six terminal devices, namely terminal device 1, terminal device 2, terminal device 3, terminal device 4, terminal device 5, and terminal device 6, etc. In Figure 1a the example shown, terminal device 1 is taken as a smart tea cup, terminal device 2 as a smart air conditioner, terminal device 3 as a smart fuel dispenser, terminal device 4 as a means of transportation, terminal device 5 as a mobile phone, and terminal device 6 as a printer for illustration.

[0154] As shown Figure 1a in the figure, the AI configuration information sending entity may be a network device. The AI configuration information receiving entities may be terminal devices 1 - 6. At this time, the network device and terminal devices 1 - 6 form a communication system. In this communication system, terminal devices 1 - 6 can send data to the network device, and the network device needs to receive the data sent by terminal devices 1 - 6. At the same time, the network device can send configuration information to terminal devices 1 - 6.

[0155] Exemplarily, in Figure 1a the figure, terminal devices 4 - 6 can also form a communication system. Among them, terminal device 5 acts as a network device, that is, the AI configuration information sending entity; terminal devices 4 and 6 act as terminal devices, that is, the AI configuration information receiving entities. For example, in a vehicle - to - everything (V2X) system, terminal device 5 sends AI configuration information to terminal devices 4 and 6 respectively, and receives the data sent by terminal devices 4 and 6; correspondingly, terminal devices 4 and 6 receive the AI configuration information sent by terminal device 5 and send data to terminal device 5.

[0156] Taking Figure 1a the communication system shown in the figure as an example, in addition to performing communication - related services, different devices (including between network devices, between network devices and terminal devices, and / or between terminal devices) may also perform AI - related services.

[0157] As Figure 1b shown in the figure, taking the network device as a base station as an example, the base station can perform communication - related services and AI - related services with one or more terminal devices, and different terminal devices can also perform communication - related services and AI - related services.

[0158] As Figure 1c shown in the figure, taking the terminal devices including a TV and a mobile phone as an example, the TV and the mobile phone can also perform communication - related services and AI - related services.

[0159] The technical solution provided by this application can be applied to wireless communication systems (such as Figure 1a , Figure 1b or Figure 1cIn the system shown, for example, an AI network element can be introduced in the communication system provided in this application to implement some or all of the AI-related operations. The AI network element can also be referred to as an AI node, AI device, AI entity, AI module, AI model, or AI unit, etc. The AI network element can be built into the network elements of the communication system. For example, the AI network element can be an AI module built into: an access network device, a core network device, a cloud server, or a network management (operation, administration and maintenance, OAM) to implement AI-related functions. The OAM can be the network management of the core network device and / or the network management of the access network device. Alternatively, the AI network element can also be an independently set network element in the communication system. Optionally, an AI entity can also be included in the terminal or the chip built into the terminal to implement AI-related functions.

[0160] Next, the artificial intelligence (AI) that may be involved in this application will be briefly introduced.

[0161] Artificial intelligence (AI) can enable machines to have human intelligence. For example, it can enable machines to apply computer software and hardware to simulate certain intelligent behaviors of humans. To achieve artificial intelligence, machine learning methods can be used. In machine learning methods, the machine learns (or trains) a model using training data. This model represents the mapping from input to output. The learned model can be used for inference (or prediction), that is, the model can be used to predict the output corresponding to a given input. Among them, this output can also be referred to as the inference result (or prediction result).

[0162] Machine learning can include supervised learning, unsupervised learning, and reinforcement learning. Among them, unsupervised learning can also be referred to as non-supervised learning.

[0163] Supervised learning is based on the collected sample values and sample labels, uses machine learning algorithms to learn the mapping relationship from sample values to sample labels, and uses an AI model to express the learned mapping relationship. The process of training a machine learning model is the process of learning this mapping relationship. During the training process, the sample values are input into the model to obtain the predicted values of the model, and the model parameters are optimized by calculating the error between the predicted values of the model and the sample labels (ideal values). After the mapping relationship learning is completed, the learned mapping can be used to predict new sample labels. The mapping relationship learned by supervised learning can include linear mapping or non-linear mapping. According to the type of label, the learning tasks can be divided into classification tasks and regression tasks.

[0164] Unsupervised learning discovers the intrinsic patterns of samples by itself using algorithms based on the collected sample values. In unsupervised learning, there is a type of algorithm that uses the samples themselves as the supervision signal, that is, the model learns the mapping relationship from samples to samples, which is called self-supervised learning. During training, the model parameters are optimized by calculating the error between the predicted value of the model and the samples themselves. Self-supervised learning can be used in applications such as signal compression and decompression recovery. Common algorithms include autoencoders and generative adversarial networks, etc.

[0165] Reinforcement learning is different from supervised learning. It is a type of algorithm that learns strategies to solve problems by interacting with the environment. Different from supervised and unsupervised learning, there is no clear "correct" action label data in reinforcement learning problems. The algorithm needs to interact with the environment to obtain the reward signal feedback from the environment, and then adjust the decision-making actions to obtain a larger reward signal value. For example, in downlink power control, the reinforcement learning model adjusts the downlink transmission power of each user according to the total system throughput rate fed back by the wireless network, and then expects to obtain a higher system throughput rate. The goal of reinforcement learning is also to learn the mapping relationship between the environmental state and the optimal (e.g., the best) decision-making actions. However, because the "correct action" labels cannot be obtained in advance, the network cannot be optimized by calculating the error between the action and the "correct action". The training of reinforcement learning is achieved through iterative interactions with the environment.

[0166] Neural network (NN) is a specific model in machine learning technology. According to the universal approximation theorem, neural networks can theoretically approximate any continuous function, enabling neural networks to have the ability to learn any mapping. Traditional communication systems need to rely on rich expert knowledge to design communication modules, while deep learning communication systems based on neural networks can automatically discover the implicit pattern structures from large datasets, establish the mapping relationship between data, and obtain better performance than traditional modeling methods.

[0167] The idea of neural networks comes from the neuron structure of the brain tissue. For example, each neuron performs a weighted sum operation on its input values and outputs the operation result through an activation function.

[0168] As Figure 2a shown, it is a schematic diagram of a neuron structure. Assume that the input of the neuron is x = [x 0 , x 1 , …, x n , and the weights corresponding to each input are w = [w, w 1 , …, w n , where n is a positive integer, and w i and x i can be various possible types such as decimals, integers (e.g., 0, positive integers, or negative integers, etc.), or complex numbers. wi As x i The weight value is used to weight x i For weighted summation. The bias for weighted summation of the input value is, for example, b. The form of the activation function can be various. Assuming the activation function of a neuron is: y = f(z) = max(0, z), then the output of this neuron is: For another example, the activation function of a neuron is: y = f(z) = z, then the output of this neuron is: Among them, b can be various possible types such as decimals, integers (such as 0, positive integers or negative integers), or complex numbers. The activation functions of different neurons in the neural network can be the same or different.

[0169] In addition, a neural network generally includes multiple layers, and each layer can include one or more neurons. By increasing the depth and / or width of the neural network, the expression ability of the neural network can be improved, providing a more powerful information extraction and abstract modeling ability for complex systems. Among them, the depth of the neural network can refer to the number of layers included in the neural network, and the number of neurons included in each layer can be called the width of this layer. In one 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 passes the processing result to the output layer, and the output layer obtains 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 passes the processing result to the intermediate hidden layer. The hidden layer calculates the received processing result to obtain a calculation result. The hidden layer passes the calculation result to the output layer or the next adjacent hidden layer, and finally the output layer obtains the output result of the neural network. Among them, a neural network can include one hidden layer, or include multiple successively connected hidden layers, without limitation.

[0170] The neural network is, for example, a deep neural network (DNN). According to the construction method of the network, DNN can include a feedforward neural network (FNN), a convolutional neural network (CNN) and a recurrent neural network (RNN).

[0171] Figure 2b It is a schematic diagram of an FNN network. The characteristic of the FNN network is that neurons between adjacent layers are completely connected in pairs. This characteristic makes FNN usually require a large amount of storage space and result in a high computational complexity.

[0172] A CNN is a neural network specifically designed to process data with a similar grid structure. For example, time series data (discretely sampled along the time axis) and image data (two-dimensionally discretely sampled) can both be considered data with a similar grid structure. Instead of using all the input information for computation at once, a CNN uses a window of a fixed size to intercept partial information for convolution operations, which greatly reduces the computational amount of model parameters. Additionally, depending on the type of information intercepted by the window (such as people and objects in the same picture being different types of information), different convolution kernels can be used for each window, enabling the CNN to better extract the features of the input data.

[0173] An RNN is a type of DNN network that utilizes feedback time series information. Its input includes the new input value at the current moment and its own output value at the previous moment. RNNs are suitable for obtaining sequential features that are correlated over time and are particularly applicable to applications such as speech recognition and channel coding and decoding.

[0174] During the model training process of the above-mentioned machine learning, a loss function can be defined. The loss function describes the gap or difference between the output value of the model and the ideal target value. The loss function can be embodied in various forms, and there is no restriction on the specific form of the loss function. The model training process can be regarded as the following process: by adjusting some or all of the parameters of the model, the value of the loss function is made less than the threshold value or meets the target requirements.

[0175] The model can also be called an AI model, a rule, or other names, etc. An AI model can be considered as a specific method for implementing AI functions. An AI model characterizes the mapping relationship or function between the input and output of the model. AI functions can include one or more of the following: data collection, model training (or model learning), model information publishing, model inference (or also called model reasoning, reasoning, or prediction, etc.), model monitoring or model verification, or the publishing of inference results, etc. AI functions can also be called AI (related) operations or AI-related functions.

[0176] Next, the implementation process of the neural network will be described exemplarily in conjunction with the accompanying drawings.

[0177] 1. Fully connected neural network, also known as a multilayer perceptron (MLP).

[0178] As Figure 2c shown, an MLP contains an input layer (on the left), an output layer (on the right), and multiple hidden layers (in the middle). Among them, each layer of the MLP contains several nodes, called neurons. Among them, the neurons in adjacent layers are pairwise connected.

[0179] Optionally, considering the neurons in adjacent layers, the output h of the neurons in the next layer is the weighted sum of all the neurons x in the previous layer connected to it and passes through an activation function, which can be expressed as:

[0180] h = f(wx + b).

[0181] Among them, w is the weight matrix, b is the bias vector, and f is the activation function.

[0182] Further optionally, the output of the neural network can be recursively expressed as:

[0183] y = f n (w n f n-1 (…)+b n ).

[0184] Among them, n is the index of the neural network layer, 1 <= n <= N, where N is the total number of layers of the neural network.

[0185] In other words, the neural network can be understood as a mapping relationship from the input data set to the output data set. Usually, the neural network is randomly initialized, and the process of obtaining this mapping relationship from the existing data with random w and b is called the training of the neural network.

[0186] Optionally, the specific training method is to evaluate the output result of the neural network using a loss function.

[0187] As Figure 2d shown, the error can be backpropagated, and the neural network parameters (including w and b) can be iteratively optimized by the method of gradient descent until the loss function reaches the minimum value, that is, Figure 2d the "preferred point (such as the optimal point)" in. Figure 2d It can be understood that the neural network parameters corresponding to the "preferred point (such as the optimal point)" in can be used as the neural network parameters in the trained AI model information.

[0188] Further optionally, the process of gradient descent can be expressed as:

[0189]

[0190] Among them, θ is the parameter to be optimized (including w and b), L is the loss function, η is the learning rate, which controls the step size of gradient descent, represents the derivative operation, represents the derivative of L with respect to θ.

[0191] Further optionally, the process of backpropagation utilizes the chain rule of partial derivatives.

[0192] AsFigure 2e As shown, the gradient of the parameters of the previous layer can be recursively calculated from the gradient of the parameters of the subsequent layer, which can be expressed as:

[0193]

[0194] where w ij is the weight connecting node j to node i, and s i is the weighted sum of inputs on node i.

[0195] 2. Federated Learning (FL).

[0196] The concept of federated learning effectively solves the dilemmas faced by the current development of artificial intelligence. On the premise of fully ensuring the privacy and security of user data, it enables various edge devices and the central server to cooperate efficiently to complete the model learning task.

[0197] As Figure 2f shown, the FL architecture is the most widely used training architecture in the current FL field. The FedAvg algorithm is the basic algorithm of FL, and its algorithm process is roughly as follows:

[0198] (1) The central server initializes the model to be trained and broadcasts it to all client devices.

[0199] (2) In the t-th round (t ∈ [1, T]), client k ∈ [1, K] trains the received global model on the local dataset for E epochs to obtain the local training result and reports it to the central node.

[0200] (3) The central node aggregates the local training results from all (or part of) the clients. Suppose the set of clients uploading local models in the t-th round is The central server will calculate the weighted average with the sample numbers of the corresponding clients as weights to obtain a new global model. The specific update rule is After that, the central server broadcasts the latest version of the global model to all client devices for a new round of training.

[0201] (4) Repeat steps (2) and (3) until the model finally converges or the number of training rounds reaches the upper limit.

[0202] In addition to reporting the local model the local gradients of the training can also be reported. The central node will average the local gradients and update the global model according to the direction of this average gradient.

[0203] As can be seen, in the FL framework, the dataset exists at the distributed nodes. That is, the distributed nodes collect the local datasets, perform local training, and report the local results (models or gradients) obtained from the training to the central node. The central node itself does not have a dataset and is only responsible for fusing the training results of the distributed nodes to obtain the global model and distributing it to the distributed nodes.

[0204] 3. Decentralized learning. Different from federated learning, another distributed learning architecture is decentralized learning.

[0205] As Figure 2g shown, consider a fully distributed system without a central node. The design objective f(x) of the decentralized learning system is generally the mean of the objective functions f i (x) of each node, that is where n is the number of distributed nodes, x is the parameter to be optimized, and in machine learning, x is the parameter of the machine learning (such as neural network) model. Each node uses the local data and the local objective f i (x) to calculate the local gradient and then sends it to the neighboring nodes that can communicate. After any node receives the gradient information sent by its neighboring nodes, it can update the parameter x of the local model according to the following formula:

[0206]

[0207] where represents the parameter of the local model after the (k + 1)th (k is a natural number) update in the i-th node, represents the parameter of the local model after the k-th update in the i-th node (if k is 0, it means is the parameter of the local model of the i-th node that has not participated in the update), α k represents the tuning coefficient, N i is the set of neighboring nodes of node i, and |N i | represents the number of elements in the set of neighboring nodes of node i, that is, the number of neighboring nodes of node i. Through the information interaction between nodes, the decentralized learning system will finally learn a unified model.

[0208] 4. Split learning.

[0209] As Figure 2h shown, in split learning, the complete neural network model is divided into two parts (i.e., two sub-networks), and one part is deployed at the distributed nodes (for example Figure 2hOn nodes 1, 2, and 3), and the other part is deployed on the central node. The place where the complete neural network is split is called the "split layer". During forward inference, the distributed nodes input local data into the local sub-network. When inferring to the split layer, the results Fk of the split layer (such as F1 / F2 / F3 in the figure) are sent to the central node through the communication link. The central node inputs the received Fk into another sub-network deployed on itself and continues forward inference to obtain the final inference result. During the gradient backpropagation of model training, the gradient is passed back from the sub-network of the central node to the split layer to obtain the backpropagation result Gk (such as G1 / G2 / G3 in the figure). Then the central node sends Gk to the distributed nodes and continues the gradient backpropagation on the sub-networks of the distributed nodes.

[0210] Optionally, in split learning, the models deployed on different distributed nodes can be the same or different, which can be determined according to the own requirements and capabilities of different distributed nodes and are not limited here.

[0211] Optionally, a distributed node can also send local model-related parameters to other distributed nodes. For example, Figure 2h node 1 in can send the local model-related parameters to node 2 and node 3 respectively. In this example, node 1 first trains the local model (denoted as model 1). Optionally, node 1 can send the model-related parameters of model 1 to other nodes so that other nodes can continue training based on model 1 and can obtain the local models of other nodes faster.

[0212] It can be seen that during the forward inference and gradient backpropagation processes of split learning, it may involve a distributed node and a central node. The sub-network on the trained distributed node can be saved locally on the distributed node or on a specific model storage server. When a new distributed node joins the learning system, it can first download the trained sub-network of the distributed node and then use local data for further training.

[0213] The technical solution provided by this application can be applied to a wireless communication system (such as Figure 1a or Figure 1b or Figure 1c the system shown). In a wireless communication system, communication nodes generally have signal transceiver capabilities and computing capabilities. Taking a network device with computing capabilities as an example, the computing capabilities of the network device mainly provide computing power support for the signal transceiver capabilities (for example: performing sending and receiving processing on signals) to achieve the communication tasks between the network device and other communication nodes.

[0214] However, in a communication network, in addition to processing the communication signals in the communication network, communication nodes may also need to take into account AI-related processing. Therefore, how to achieve the integration of AI-related processing and communication networks is a technical problem that needs to be solved urgently.

[0215] In order to solve the above problems, the present application provides a communication method and related equipment, which are used to enable communication nodes to collect artificial intelligence (AI) data. The following will be described in detail with reference to the accompanying drawings.

[0216] See also Figure 3 , is a schematic diagram of an implementation of the communication method provided in this application, and the method includes the following steps.

[0217] It should be noted that in Figure 3 In the example, the first node and the control node are used as the execution subjects of the interaction diagram to illustrate the method, but the present application does not limit the execution subjects of the interaction diagram. Figure 3 and later Figure 6 In the method, the execution subject can be replaced by a chip, a chip system, a processor, a logic module or software in a communication device. The first node can be a terminal device and the control node can be a network device, or the first node and the control node are both terminal devices (for example, the method can be applied to the communication process of different terminal devices in a sidelink communication scenario).

[0218] S301. The control node sends first configuration information, and correspondingly, the first node receives the first configuration information, wherein the first configuration information is used to configure AI data collection.

[0219] S302. The first node collects data based on the first configuration information to obtain first AI data, wherein the first AI data is used for model processing of the first AI model.

[0220] In this application, terms such as data collection, data collection, data collection, data acquisition, and data capture can be used interchangeably.

[0221] In this application, terms such as AI model, neural network model, AI neural network model, machine learning model, and AI processing model can be used interchangeably.

[0222] Optionally, wireless communication signals (such as the transceiver of configuration information of communication resources, the transceiver of reference signals, etc.) can be transmitted between different communication nodes (such as the first node, the control node, and other nodes mentioned later, including the second node, the central node, etc.). The AI models (such as the first AI model, the second AI model mentioned later, etc.) involved in the embodiments of the present application can be used to process the wireless communication signals (including at least one of management, configuration, update, and optimization). For example, the AI model can include one or more of an AI model for modulation and / or demodulation, an AI model for channel prediction, an AI model for beam management, an AI model for assisting positioning, an AI model for channel compression, an AI model for resource scheduling, an AI model for mobility management, an AI model for load balancing, an AI model for network energy saving, and an AI model for replacing one or more modules in a transmitter and / or a receiver. Alternatively, the AI models involved in the embodiments of the present application can also be AI models for other AI tasks, such as an AI model for image recognition, an AI model for natural language processing, an AI model for computer vision, etc.

[0223] In Figure 3 In a possible implementation of the shown solution, the first node is one of the N distributed nodes, and N is an integer greater than or equal to 1.

[0224] As an implementation example, as Figure 4a shown, taking the value of N greater than 2 as an example, the N nodes can include Figure 4a nodes 1, 2... N among them, and the control node can communicate with the N distributed nodes. Among them, the first node can be any one of the N nodes. In other words, any one of the N distributed nodes can execute the method executed by the first node, so that the N distributed nodes can all be used as nodes for data collection to achieve AI data collection in a distributed scenario.

[0225] In a possible implementation, as Figure 4b shown in the example, in addition to communicating with the control node, the N distributed nodes can also communicate with the data receiving node.

[0226] As Figure 4c shown in the example, when the control node and the data receiving node are different nodes, compared with Figure 3 the shown implementation process, after step S302, the first node can also send the first AI data to the data receiving node in step A, and subsequently, the data receiving node can perform model processing on the first AI model based on the first AI data.

[0227] In a possible implementation, as Figure 4d shown in the example, in addition to communicating with the control node, the N distributed nodes can also communicate with the data receiving node. Moreover, when the control node and the data receiving node are the same node, the functions of the control node (such as determining and issuing configuration information, etc.) and the data receiving node (such as receiving AI data, etc.) are implemented through the same node.

[0228] Similarly, as Figure 4e shown in the example, compared with Figure 3 the implementation process shown, after step S302, the first node can also send the first AI data to the control node (data receiving node) in step B, and subsequently, the control node (data receiving node) can perform model processing on the first AI model based on the first AI data.

[0229] It should be noted that the data receiving node can implement the process of performing model processing on the first AI model in various ways. For example, when the first AI model is deployed on the data receiving node, the data receiving node can perform model processing on the first AI data locally based on the first AI model. Another example is that when the first AI model is deployed on other nodes (such as one or more of the N distributed nodes, or Figure 4b / Figure 4d other nodes different from the N distributed nodes not shown in the figure (such as the central node)), the data receiving node can send the first AI data to the other node so that the other node can subsequently perform model processing on the first AI data based on the first AI model.

[0230] Optionally, the above model processing includes at least one of model training, model inference, and model monitoring. Correspondingly, the first AI data sent by the first node in step A or step B includes at least one of the AI data used in the model training stage of the first AI model, the AI data used in the model inference stage of the first AI model, and the AI data used in the model monitoring stage of the first AI model.

[0231] In one implementation example, the AI data used in the model training stage of the first AI model included in the first AI data may include at least one of the input data, feature data, and label data for training the first AI model.

[0232] In another implementation example, the AI data used in the model inference stage of the first AI model included in the first AI data may include at least one of the input data, feature data, and inference result data for inference of the first AI model.

[0233] In another implementation example, the AI data used in the model monitoring phase of the first AI model included in the first AI data may include at least one of input data, feature data, label data, inference result data, AI model performance data, and communication performance data for monitoring the first AI model.

[0234] Optionally, the feature data may indicate intermediate data, intermediate results, etc. of the inference of the AI model.

[0235] Optionally, the AI model performance data may refer to learning performance, such as accuracy, inference latency, inference complexity, etc.

[0236] Optionally, the communication performance data may refer to communication system performance, such as throughput, packet loss rate, latency, etc.

[0237] It should be noted that for a distributed node, in the process of the distributed node collecting AI data based on configuration information (for example, the process of the first node collecting the first AI data based on the first configuration information in step S302), the distributed node can collect AI data based on the following multiple methods.

[0238] Method 1. The distributed node collects AI data based on the communication process.

[0239] Exemplarily, in Method 1, the distributed node is used as a communication node, and the involved communication process may include modulation and / or demodulation of signals, measurement of reference signals, transceiver of sensing signals, etc. Correspondingly, the AI data collected by the distributed node based on the communication process may include one or more of the modulation and / or demodulation results of signals, the measurement results of reference signals, and the sensing results of sensing signals.

[0240] Method 2. The distributed node can collect AI data based on the data involved in the model processing of the AI model deployed locally.

[0241] Exemplarily, in Method 2, the AI data collected by the distributed node may include one or more of the data for model training of the local AI model, the data for model inference of the local AI model, and the data for model monitoring of the local AI model.

[0242] It can be understood that from the above description, for the first node, the AI model deployed locally on the first node may include the above-mentioned first AI model, or may include other AI models different from the first AI model, which is not limited here.

[0243] It is understandable that the AI data collected by the above-mentioned Method 1 and Method 2 may include the same parts, that is, it is possible that the data obtained by Method 1 is the same as the data obtained by Method 2. For example, in the case where the AI model locally deployed at a certain distributed node is an AI model for signal modulation and / or demodulation, the communication signals transmitted and received by this distributed node can be the data interacted in the communication process of Method 1, or can be the data involved in the model processing of Method 2.

[0244] In Figure 3 In a possible implementation manner of the shown solution, after step S302, the method further includes: the first node sends the data of the first node to other nodes among the N distributed nodes, and the data of the first node is used for data collection of other nodes. Specifically, for the N distributed nodes, part or all of the AI data collected by one of the distributed nodes can be determined by the data sent by other distributed nodes. Correspondingly, the first node can also send the data of the first node to other nodes among the N distributed nodes. For example, when the first node is Figure 4a / Figure 4b / Figure 4d Node 1 in, the other nodes can be one or more nodes among Node 2 to Node N, so that other nodes can implement data collection based on the data of the first node. Exemplarily, the other nodes can use the data of the first node received as the data in Method 1 and / or Method 2 to determine the AI data collected by the other nodes.

[0245] Optionally, the data of the first node can include part or all of the first AI data.

[0246] Optionally, the data of the first node can include the communication data of the first node, such as reference signals, positioning data, etc.

[0247] Optionally, the data of the first node can include the model data of the local AI model of the first node, such as at least one of the AI data used in the model training stage of the local AI model, the AI data used in the model inference stage of the local AI model, and the AI data used in the model monitoring stage of the local AI model.

[0248] In Figure 3In a possible implementation of the illustrated scheme, before step S302, the method further includes: the first node receives data from other nodes among the N distributed nodes, and the data of the other nodes is used to determine part or all of the first AI data. Similarly, for the N distributed nodes, part or all of the AI ​​data collected by one of the distributed nodes can be determined by data sent by other distributed nodes. Accordingly, the first node can also receive data from other nodes among the N distributed nodes (for example, second data from the second node) so that the first node can realize data collection based on the data of the other nodes. Exemplarily, the first node can use the received data from other nodes as data in method 1 and / or method 2 to determine the first AI data collected by the first node.

[0249] Optionally, as described above, the first AI model may be deployed on a data receiving node or other nodes, and the second AI model may be deployed on other nodes among the N distributed nodes. The first AI model and the second AI model may be associated AI models.

[0250] For example, these two AI models can be Figure 2f The AI ​​model deployed on the central node in the central learning scenario and the AI ​​model deployed on the distributed nodes.

[0251] For example, these two AI models can be Figure 2f An AI model deployed on any two adjacent or non-adjacent distributed nodes in a central learning scenario.

[0252] For example, these two AI models can be Figure 2g An AI model deployed on any two adjacent or non-adjacent distributed nodes in distributed learning.

[0253] For example, these two AI models can be Figure 2h The AI ​​model deployed on the central node and the AI ​​model deployed on the distributed node in the segmentation learning.

[0254] For example, these two AI models can be Figure 2h An AI model deployed on any two adjacent or non-adjacent distributed nodes in segmentation learning.

[0255] Optionally, the data of the other node may include communication data of the other node, such as reference signals, positioning data, etc.

[0256] Optionally, the data of the other node may include model data of the local AI model of the other node, such as at least one of the AI data used in the model training phase of the local AI model, the AI data used in the model inference phase of the local AI model, and the AI data used in the model monitoring phase of the local AI model. Exemplarily, taking the other node as the second node, the local AI model of the second node may be denoted as the second AI model, and the model data of the second AI model may be denoted as the second AI data. Among them, the implementation of the second AI data is similar to that of the first AI data. The second AI data received by the first node may include at least one of the AI data used in the model training phase of the second AI model, the AI data used in the model inference phase of the second AI model, and the AI data used in the model monitoring phase of the second AI model.

[0257] In one implementation example, the AI data used in the model training phase of the second AI model included in the second AI data includes at least one of the input data, feature data, and label data used for training the second AI model.

[0258] In another implementation example, the AI data used in the model inference phase of the second AI model included in the second AI data includes at least one of the input data, feature data, and inference result data used for inference of the second AI model.

[0259] In another implementation example, the AI data used in the model monitoring phase of the second AI model included in the second AI data includes at least one of the input data, feature data, label data, inference result data, AI model performance data, and communication performance data used for monitoring the second AI model.

[0260] As described above Figures 2f to 2h As can be seen from the description of the AI processing architecture shown above, in the interaction process between different distributed nodes, it is possible that intermediate nodes are involved, or it is possible that no intermediate nodes are involved. More implementation examples will be described below.

[0261] In a possible implementation, in step S301, the first node can receive the first configuration information in multiple ways. For example, the process of the first node receiving the first configuration information includes: the first node receives the first configuration information from the control node, where the control node is used to control the data collection of the N distributed nodes; or, the first node receives the first configuration information from the control node through the central node; or, the first node receives the first configuration information from the second node, where the second node is a node different from the first node among the N distributed nodes, and N is greater than 1. Specifically, the first node can receive the first configuration information in the above multiple ways. The first node is one of the N distributed nodes. In this way, the distributed nodes can receive the first configuration information in multiple different scenarios, and the flexibility of the solution implementation is improved.

[0262] Optionally, the control node and the central node can be the same node, or, the control node and the central node can be different logical nodes in the physical nodes, or, the control node and the central node can be two independent different nodes.

[0263] As an implementation example, the first configuration information received by the first node is the configuration information corresponding to the first node among the M configuration information, and the M configuration information is at least used to configure the AI data collection of M distributed nodes among the N distributed nodes, where M is less than or equal to N; the method further includes: the first node sends at least one of the M configuration information to at least one of the M distributed nodes. Specifically, the first node can receive the M configuration information and determine the first configuration information from the M configuration information. Among them, the M configuration information is at least used to configure the AI data collection of M distributed nodes among the N distributed nodes. Correspondingly, the first node can send at least one of the M configuration information to other nodes among the M distributed nodes, so that other distributed nodes can obtain the corresponding configuration information and perform AI data collection.

[0264] Optionally, the first node sends at least one of the M configuration information to at least one of the M distributed nodes, so that each of the M distributed nodes can obtain its corresponding configuration information, so that the M distributed nodes can implement AI data collection based on their corresponding configuration information. Further optionally, during the sending process, the first node can send the configuration information corresponding to each of the M distributed nodes to each of the M distributed nodes, or the first node can send the M configuration information to each of the M distributed nodes, or the first node can send the M configuration information to some of the M distributed nodes, and the some distributed nodes send the configuration information corresponding to the other some distributed nodes to the other some distributed nodes, or in other ways to enable each of the M distributed nodes to obtain its corresponding configuration information, which is not limited here.

[0265] Optionally, among the N distributed nodes, the configuration information corresponding to different distributed nodes may be different from each other. Therefore, the M configuration information are respectively used to configure the AI data collection of M of the N distributed nodes.

[0266] Optionally, among the N distributed nodes, the configuration information corresponding to different distributed nodes may be the same. Therefore, in addition to the M configuration information being respectively used to configure the AI data collection of M of the N distributed nodes, at least one of the M configuration information can also be used to configure the AI data collection of at least one of the other distributed nodes in the N distributed nodes except the M distributed nodes.

[0267] Exemplarily, taking the case where the configuration information corresponding to different distributed nodes may be different from each other as an example, that is, the M configuration information is respectively used to configure the AI data collection of the M distributed nodes. In other words, the first node can send the other M - 1 configuration information except the first configuration information among the M configuration information to the other M - 1 distributed nodes except the first node among the M distributed nodes. Among them, the first node sends the other M - 1 configuration information except the first configuration information among the M configuration information in multiple ways. For example, in the case where the M - 1 configuration information is respectively used for the AI data collection of M - 1 different distributed nodes, the first node can respectively send the M - 1 configuration information to the M - 1 different distributed nodes; or, the first node can send the M - 1 configuration information to the next-hop node among the M - 1 different distributed nodes, and the next-hop node will obtain the local configuration information from the M - 1 configuration information and send the other M - 2 configuration information except the local configuration information among the M - 1 configuration information to the next-hop node of the next-hop node, and so on, until the M - 1 different distributed nodes all obtain their respective configuration information.

[0268] It can be understood that in the case where the configuration information corresponding to different distributed nodes may be the same, the first node can refer to the above multiple ways to send the M - 1 configuration information.

[0269] In a possible implementation manner, after the first node obtains the first AI data in step S302, the first node can send the first AI data in multiple ways. For example, the process of the first node sending the first AI data includes: the first node sends the first AI data to a data receiving node, and the data receiving node is used for AI data collection; or, the first node sends the first AI data to a central node. Specifically, the first node can send the first AI data in the above multiple ways, and the first node is one of the N distributed nodes. In this way, the distributed nodes can send the collected AI data in multiple different scenarios and improve the flexibility of the solution implementation.

[0270] Optionally, the control node and the data receiving node can be the same node, or, the control node and the data receiving node can be different logical nodes in a physical node, or, the control node and the data receiving node can be two independent different nodes.

[0271] From the above implementation process, it can be seen that the distributed nodes can receive the configuration information and send the AI data in multiple ways. The following will be described exemplarily in combination with Figures 5a to 5c the scenarios shown. It should be understood that in the following Figures 5a to 5c shown examples, taking Figure 4cTaking the case where the control node and the data receiving node are the same node as an example for illustration, in actual applications, the control node and the data receiving node can be different nodes.

[0272] Implementation method A: The control node (data receiving node) communicates with N distributed nodes respectively to send the configuration information of the N distributed nodes and receive the AI data of the N distributed nodes.

[0273] As an implementation example of Implementation method A, as Figure 5a shown, when the control node (data receiving node) can communicate with the N distributed nodes, the control node (data receiving node) can communicate with each distributed node. For example, there is a direct connection link between the control node (data receiving node) and each distributed node for communication, or the control node (data receiving node) and each distributed node can communicate through one or more relay nodes.

[0274] Correspondingly, in Figure 5a , the N distributed nodes can respectively act as the first nodes, that is, the control node (data receiving node) can execute the process of sending the first configuration information N times in step S301, so that the N distributed nodes respectively receive their own configuration information, and the N distributed nodes can implement the acquisition of AI data based on their respective configuration information in step S302. In addition, as shown in step B in Figure 4e , after step S302, the N distributed nodes can send the AI data collected by themselves to the control node (data receiving node).

[0275] Implementation method B: The control node (data receiving node) communicates with one of the N distributed nodes to send the configuration information of the N distributed nodes and receive the AI data of the N distributed nodes.

[0276] As an implementation example of Implementation method B, as Figure 5b shown, when the control node (data receiving node) can communicate with one of the N distributed nodes ( Figure 5b taking node 1 as an example in

[0277] ), and different distributed nodes can communicate with each other, the control node (data receiving node) can send the configuration information of the N distributed nodes through the communication process with node 1. For example, there is a direct connection link between the control node (data receiving node) and node 1 for communication, or the control node (data receiving node) and node 1 can communicate through one or more relay nodes. Figure 5bAmong them, node 1 serves as the first node and can receive K configuration information containing the first configuration information in step S301.

[0278] Optionally, among the N distributed nodes, the configuration information corresponding to different distributed nodes may be different from each other. For this reason, the values of K and N can be equal, that is, the K configuration information can be respectively used to configure the AI data collection of different nodes in the N distributed nodes.

[0279] Optionally, among the N distributed nodes, the configuration information corresponding to different distributed nodes may be the same. For this reason, K can be less than N, that is, at least one of the K configuration information is used to configure the AI data collection of at least two nodes in the N distributed nodes. Correspondingly, the configuration information of the AI data collection of the at least two nodes is the same.

[0280] Exemplarily, taking the case where the K configuration information can be respectively used to configure the AI data collection of different nodes in the N distributed nodes as an example, that is, K is equal to N. For the convenience of understanding, the following will record the K configuration information as N configuration information. When N is greater than 1, in the case where there are communication links (such as the dotted line in Figure 5b ) between node 1 and the other N - 1 nodes, node 1 can respectively send the other N - 1 configuration information except the first configuration information in the N configuration information to the other N - 1 nodes, so that all N distributed nodes can obtain their respective configuration information and perform data collection based on the configuration information. Alternatively, node 1 can send the other N - 1 configuration information except the first configuration information in the N configuration information to the adjacent node (such as node 2). Similarly, after node 2 obtains its corresponding configuration information (such as the second configuration information) from the N - 1 configuration information, node 2 can also send the other N - 2 configuration information except the first configuration information and the second configuration information in the N configuration information to the adjacent node, and so on, until all N distributed nodes can obtain their respective configuration information and perform data collection based on the configuration information.

[0281] After that, in Figure 5b , nodes 2 to N can refer to the reverse transmission process of the above process and send the AI data collected by themselves to node 1. Subsequently, node 1 can send the AI data collected by the N distributed nodes to the control node (data receiving node) through one or more sending processes. In this way, the control node (data receiving node) can realize the distribution of the configuration information of the N distributed nodes and the reception of the AI data collected by the N distributed nodes.

[0282] It should be noted that the implementation processes of implementation method A and implementation method B can be combined with each other. For example, M (M is a positive integer) nodes among N distributed nodes communicate in the manner of implementation method B, and the other N - M nodes communicate in the manner of implementation method A. In other words, for these M nodes, the communication process between the control node (data receiving node) and one of the M nodes can be used to send the configuration information of the M distributed nodes and receive the AI data of the M distributed nodes; for the N - M nodes, the control node (data receiving node) communicates with the N - M distributed nodes respectively to send the configuration information of the N - M distributed nodes and receive the AI data of the N - M distributed nodes. The specific implementation process can refer to the implementation processes of the above implementation method A and implementation method B.

[0283] Implementation method C: The control node (data receiving node) communicates with the central node to send the configuration information of N distributed nodes and receive the AI data of N distributed nodes.

[0284] As an implementation example of implementation method C, as Figure 5c shown, when the control node (data receiving node) can communicate with the central node, the control node (data receiving node) can send the configuration information of N distributed nodes through the communication process with the central node. For example, there is a direct connection link between the control node (data receiving node) and the central node for communication, or the control node (data receiving node) and the central node can communicate through one or more relay nodes.

[0285] Correspondingly, in Figure 5c , any one of the N distributed nodes can be used as the first node. In step S301, the first configuration information is received through the central node, so that all N distributed nodes can obtain their respective configuration information, and in step S302, data collection is performed based on the configuration information.

[0286] After that, in Figure 5c , any one of the N distributed nodes can be used as the first node. After step S302, the AI data collected by each node is sent to the central node, so that the central node obtains N pieces of AI data. After that, the central node can process the received N pieces of AI data (such as data screening, data merging, data redundancy removal, etc.) and then send it to the control node (data receiving node), or the central node can transparently forward the N pieces of AI data to the control node (data receiving node). Through this centralized implementation method, the control node (data receiving node) can issue the configuration information of N distributed nodes and receive the AI data collected by N distributed nodes through the central node.

[0287] Optionally, in Figure 5c , the control node can communicate with one or more of the N distributed nodes without passing through the central node. For this purpose, the process of the above-mentioned distributed node receiving configuration information can achieve the sending and receiving of configuration information through the transmission of the central node, or can also achieve the sending and receiving of configuration information without passing through the transmission of the central node (reference can be made to the implementation processes shown in the previous Figure 5a and Figure 5b ), and no limitation is imposed here. Similarly, the process of the above-mentioned distributed node sending AI data can achieve the sending and receiving of AI data through the transmission of the central node, or can also achieve the sending and receiving of AI data without passing through the transmission of the central node (reference can be made to the implementation processes shown in the previous Figure 5a and Figure 5b ), and no limitation is imposed here.

[0288] It should be noted that there may be communication links that are reachable for communication between the central node and all of the N distributed nodes, or there may be communication links that are reachable only between the central node and some of the distributed nodes, and these two implementation processes can refer to the aforementioned implementation method A, implementation method B, and the implementation processes of the previous M nodes and N - M nodes.

[0289] Optionally, in Figure 5c , the central node can be deployed on the same node as the control node and the data receiving node. In other words, the central node, the control node, and the data receiving node can be three different nodes, or any two or three of the central node, the control node, and the data receiving node can be the same node.

[0290] In a possible implementation manner, the first configuration information received by the first node in step S301 includes at least one of the following information A to information J.

[0291] Information A. The identifier (or index) of the AI model corresponding to the collected AI data.

[0292] Information B. The identifier (or index) of the AI function corresponding to the collected AI data.

[0293] Information C. The indication information indicating the model processing corresponding to the collected AI data.

[0294] Information D. The indication information indicating that the configuration method of the configuration information for AI data collection is centralized or decentralized.

[0295] Information E. The indication information indicating that the AI data collection method is centralized or decentralized.

[0296] Information F. The indication information indicating the AI data characteristics of the collected AI data.

[0297] Information G. Instruction information for AI data processing of the collected AI data.

[0298] Information H. Instruction information indicating the collection period of AI data.

[0299] Information I. Instruction information indicating the transmission information of the collected AI data.

[0300] Information J. Identification of the source node of the collected AI data.

[0301] Regarding Information A, as can be seen from the definition of the AI model above, the identification of the AI model in Information A can be used to identify (or indicate) the AI model. In other words, when the first configuration information includes Information A, the first AI data obtained by the first node for data collection based on this first configuration information includes the data of the AI model corresponding to the identification of the AI model indicated by Information A. For example, the identification of the AI model in Information A can specifically be the identification of the AI model for modulation and / or demodulation, the identification of the AI model for channel prediction, the identification of the AI model for beam management, the identification of the AI model for assisted positioning, the identification of the AI model for channel compression, the identification of the AI model for resource scheduling, the AI model for mobility management, the AI model for load balancing, the AI model for network energy saving, the identification of the AI model for replacing one or more modules in the transmitter and / or receiver, the identification of the AI model for image recognition, the identification of the AI model for natural language processing, the identification of the AI model for computer vision, etc.

[0302] Regarding Information B, as can be seen from the definition of the AI model above, the identification of the AI function in Information B can be used to identify (or indicate) the function of the AI model. In other words, when the first configuration information includes Information B, the first AI data obtained by the first node for data collection based on this first configuration information includes the AI data corresponding to the identification of the AI function indicated by Information B. For example, the identification of the AI function in Information B can specifically be the function identification for modulation and / or demodulation, the function identification for channel prediction, the function identification for beam management, the function identification for assisted positioning, the function identification for channel compression, the function identification for resource scheduling, the AI model for mobility management, the AI model for load balancing, the AI model for network energy saving, the function identification for replacing one or more modules in the transmitter and / or receiver, the function identification for image recognition, the function identification for natural language processing, the function identification for computer vision, etc.

[0303] For information C, since model processing can include at least one of model training, model inference, and model monitoring, information C can include at least one of indication information indicating that the collected AI data is for (or corresponds to) model training, indication information indicating that the collected AI data is for (or corresponds to) model inference, and indication information indicating that the collected AI data is for (or corresponds to) model monitoring. In other words, when the first configuration information includes information C, the first AI data obtained by the first node based on this first configuration information for data collection includes the AI data for at least one model processing indicated by this information C.

[0304] For information D and information E, as can be seen from the example shown above Figures 5a to 5c The data exchanged between the distributed node and the control node (data receiving node) can be transmitted through the central node or not through the central node. Therefore, through the indication of information D, it can be determined whether the configuration process of the configuration information (such as the first configuration information) involves transmission through the central node. If so, the configuration method of the configuration information indicating AI data collection indicated by information D is centralized; if not, the configuration method of the configuration information indicating AI data collection indicated by information D is decentralized. In other words, when the first configuration information includes information D, if information D indicates centralized, then the first configuration information received by the first node is configured through the central node; if information D indicates decentralized, then the first configuration information received by the first node is not configured through the central node, but through the control node.

[0305] Similarly, through the indication of information E, it can be determined whether the collection process of the AI data (such as the first AI data) involves transmission through the central node; if so, information E indicates that the AI data collection method is centralized; if not, information E indicates that the AI data collection method is decentralized. In other words, when the first configuration information includes information E, if information E indicates centralized, then after the first node obtains the first AI data based on the first configuration information for data collection, the first node sends the first AI data to the central node; if information E indicates decentralized, then after the first node obtains the first AI data based on the first configuration information for data collection, the first node sends the first AI data to the data receiving node instead of sending the first AI data to the central node.

[0306] For information F, the AI data characteristics include one or more of the quantity of AI data, sample size, collection time, collection location, and distribution. In other words, when the first configuration information includes information F, the first AI data obtained by the first node based on this first configuration information for data collection satisfies the AI data characteristics indicated by this information F.

[0307] For information G, the AI data processing includes post - processing of the output data of the AI model and / or pre - processing of the input data of the AI model, such as one or more of dimension conversion and precision conversion. In other words, when the first configuration information includes information G, the first AI data obtained by the first node based on this first configuration information for data collection satisfies the post - processing and / or pre - processing indicated by this information G.

[0308] For information H, the collection period of AI data includes that AI data is collected in a periodic manner, AI data is collected in a semi - static manner, or AI data is collected in a (dynamic) trigger manner, etc. In other words, when the first configuration information includes information H, the first node performs data collection based on the collection period indicated by this information H, and then obtains the first AI data.

[0309] For information I, the transmission information of AI data includes one or more of data structure, format, precision, dimension, and transmission resources. In other words, when the first configuration information includes information I, the first AI data obtained by the first node based on this first configuration information for data collection satisfies the transmission information indicated by this information I.

[0310] For information J, the first node can, based on the identifier of the source node of the AI data to be collected indicated by this information J, request / acquire / collect data from this source node, and determine part or all of the first AI data based on the data from this source node.

[0311] It should be understood that when the first configuration information includes at least one of the above - mentioned information A to information J, in step S301, this first configuration information can be sent through one or more messages, that is, the first node can obtain this first configuration information through the receiving process of one or more messages.

[0312] Based on Figure 3 According to the technical solution shown, after the first node receives the first configuration information in step S301, the first node can perform AI data collection based on this first configuration information in step S302 to obtain the first AI data; thereafter, the first node can send this first AI data, and subsequently, the recipient of this first AI data can perform model processing of the first AI model based on this first AI data. Thus, when the communication node in the communication system is an AI participating node, it enables the communication node to be an AI data collection node and realizes the collection of AI data.

[0313] In addition, as a communication node, after the first node sends the first AI data, it enables the recipient of this first AI data to perform model processing of the AI model based on the AI data collected by this communication node.

[0314] Please refer toFigure 6 , this application also provides a communication architecture, which can be used for data acquisition.

[0315] In Figure 6 , the communication architecture at least includes a data storage module and a data transmission module. Among them, the data transmission module can be used to transmit data between nodes (including between distributed nodes, between a distributed node and a central node). The data storage module can be used to store the data received from other nodes.

[0316] Exemplarily, the data storage module can include the data receiving node in any of the above embodiments, or the data receiving node in any of the above embodiments can be used to execute the process performed by the data storage model.

[0317] Exemplarily, the data transmission module can include the distributed nodes (such as N distributed nodes) and / or the central node in any of the above embodiments, or the distributed nodes (such as N distributed nodes) and / or the central node in any of the above embodiments can be used to execute the process performed by the data transmission model.

[0318] Optionally, the communication architecture can further include a data measurement module. Among them, the data measurement module can be used to obtain measurement data through measurement (including measurement based on a reference signal, measurement through a sensing function). Correspondingly, the data storage module can also be used to store the data obtained by this node through measurement.

[0319] Exemplarily, the data measurement module can include the nodes (such as N distributed nodes, a control node, a central node, a data receiving node) in any of the above embodiments, or the nodes (such as N distributed nodes, a control node, a central node, a data receiving node) in any of the above embodiments can be used to execute the process performed by the data measurement module.

[0320] Optionally, the communication architecture can further include a data usage module. Among them, the data usage module can obtain data for model training, model inference, and model monitoring from the data storage, and store the data that may be generated during model training, model inference, and model monitoring again. Correspondingly, the data storage module can also be used to store the data generated during the data usage process.

[0321] Exemplarily, the data usage module can include the nodes (such as N distributed nodes, a control node, a central node, a data receiving node) with an AI model deployed in any of the above embodiments, or the nodes (such as N distributed nodes, a control node, a central node, a data receiving node) in any of the above embodiments can be used to execute the process performed by the data usage module.

[0322] Please refer toFigure 7 , an embodiment of the present application provides a communication device 700. The communication device 700 can implement the functions of the first node (or control node) in the above method embodiment, and thus can also achieve the beneficial effects of the above method embodiment. In the embodiment of the present application, the communication device 700 can be the first node (or control node), or an integrated circuit or component inside the first node (or control node), such as a chip.

[0323] It should be noted that the transceiver unit 702 can include a sending unit and a receiving unit, which are respectively used for sending and receiving.

[0324] In a possible implementation manner, when the device 700 is used to execute the method performed by the first node in the foregoing embodiment, the device 700 includes a processing unit 701 and a transceiver unit 702; the transceiver unit 702 is used to receive first configuration information, and the first configuration information is used to configure AI data collection; the processing unit 701 sends first AI data, and the first AI data is collected based on the first configuration information; wherein, the first AI data is used for model processing of the first AI model.

[0325] In a possible implementation manner, when the device 700 is used to execute the method performed by the control node in the foregoing embodiment, the device 700 includes a processing unit 701 and a transceiver unit 702; the processing unit 701 is used to determine first configuration information, and the first configuration information is used for AI data collection; the transceiver unit 702 is used to send the first configuration information.

[0326] It should be noted that for the content such as the information execution process of the units of the above communication device 700, please refer to the description in the method embodiment shown in the foregoing of the present application, and details are not described herein again.

[0327] Please refer to Figure 8 , which is another schematic structural diagram of the communication device 800 provided by the present application. The communication device 800 includes a logic circuit 801 and an input / output interface 802. Among them, the communication device 800 can be a chip or an integrated circuit.

[0328] Among them, Figure 7 the shown transceiver unit 702 can be a communication interface, and the communication interface can be Figure 8 the input / output interface 802 in , and the input / output interface 802 can include an input interface and an output interface. Alternatively, the communication interface can also be a transceiver circuit, and the transceiver circuit can include an input interface circuit and an output interface circuit.

[0329] Optionally, the input / output interface 802 is used to receive first configuration information for configuring AI data collection; the logic circuit 801 transmits first AI data collected based on the first configuration information; wherein, the first AI data is used for model processing of the first AI model.

[0330] Optionally, the logic circuit 801 is used to determine first configuration information for AI data collection; the input / output interface 802 is used to transmit the first configuration information.

[0331] Wherein, the logic circuit 801 and the input / output interface 802 can also perform other steps executed by the first node or the control node in any embodiment and achieve corresponding beneficial effects, which will not be elaborated here.

[0332] In a possible implementation manner, Figure 7 the processing unit 701 shown can be Figure 8 the logic circuit 801 in

[0333] Optionally, the logic circuit 801 can be a processing device, and the functions of the processing device can be implemented partially or entirely by software. Among them, the functions of the processing device can be implemented partially or entirely by software.

[0334] Optionally, the processing device can include a memory and a processor. Among them, the memory is used to store a computer program, and the processor reads and executes the computer program stored in the memory to perform corresponding processing and / or steps in any method embodiment.

[0335] Optionally, the processing device can only include a processor. The memory for storing the computer program is located outside the processing device, and the processor is connected to the memory through a circuit / wire to read and execute the computer program stored in the memory. Among them, the memory and the processor can be integrated together or physically independent of each other.

[0336] Optionally, the processing device may be one or more chips, or one or more integrated circuits. For example, the processing device may be one or more field-programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), system on chips (SoCs), central processor units (CPUs), network processors (NPs), digital signal processing circuits (DSPs), microcontroller units (MCUs), programmable logic devices (PLDs), or other integrated chips, or any combination of the above chips or processors, etc.

[0337] Please refer to Figure 9 , for the communication device 900 involved in the above embodiments provided by the embodiments of the present application. The communication device 900 may specifically be the communication device serving as a terminal device in the above embodiments. Figure 9 The example shown is implemented by the terminal device (or components in the terminal device).

[0338] Among them, a possible schematic logical structure of the communication device 900 is shown. The communication device 900 may include, but is not limited to, at least one processor 901 and a communication port 902.

[0339] Among them, Figure 7 the shown transceiver unit 702 may be a communication interface, and this communication interface may be Figure 9 the communication port 902 in

[0340]

[0341] Further optionally, the device may further include at least one of a memory 903 and a bus 904. In the embodiments of the present application, the at least one processor 901 is used to control and process the actions of the communication device 900.In addition, the processor 901 may be a central processing unit, a general-purpose processor, a digital signal processor, an application specific integrated circuit, a field programmable gate array or other programmable logic device, transistor logic device, hardware component, or any combination thereof. It may implement or execute various exemplary logical blocks, modules, and circuits described in connection with the disclosure of this application. The processor may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a digital signal processor and a microprocessor, and so on. Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the systems, devices, and units described above may refer to the corresponding processes in the foregoing method embodiments, and will not be elaborated herein.

[0342] It should be noted that Figure 9 the communication device 900 shown can specifically be used to implement the steps implemented by the terminal device in the foregoing method embodiments, and achieve the technical effects corresponding to the terminal device. Figure 9 For the specific implementation manners of the communication device shown, reference can be made to the descriptions in the foregoing method embodiments, and details will not be repeated herein one by one.

[0343] Please refer to Figure 10 for the structural schematic diagram of the communication device 1000 involved in the foregoing embodiments provided in the embodiments of the present application. The communication device 1000 may specifically be the communication device acting as a network device in the foregoing embodiments. Figure 10 The example shown is implemented by a network device (or a component in the network device). Among them, the structure of the communication device may refer to Figure 10 the structure shown.

[0344] The communication device 1000 includes at least one processor 1011 and at least one network interface 1014. Further optionally, the communication device further includes at least one memory 1012, at least one transceiver 1013, and one or more antennas 1015. The processor 1011, the memory 1012, the transceiver 1013, and the network interface 1014 are connected, for example, connected by a bus. In the embodiments of the present application, this connection may include various interfaces, transmission lines, or buses, etc., and this embodiment does not make any limitations thereto. The antenna 1015 is connected to the transceiver 1013. The network interface 1014 is used to enable the communication device to communicate with other communication devices through a communication link. For example, the network interface 1014 may include a network interface between the communication device and a core network device, such as an S1 interface. The network interface may include a network interface between the communication device and other communication devices (such as other network devices or core network devices), such as an X2 or Xn interface.

[0345] Among them, Figure 7 the transceiver unit 702 shown may be a communication interface, and this communication interface may beFigure 10 The network interface 1014 therein, which may include an input interface and an output interface. Alternatively, the network interface 1014 may also be a transceiver circuit, which may include an input interface circuit and an output interface circuit.

[0346] The processor 1011 is mainly used to process communication protocols and communication data, and to control the entire communication device, execute software programs, and process the data of software programs. For example, it is used to support the communication device to perform the actions described in the embodiments. The communication device may include a baseband processor and a central processor. The baseband processor is mainly used to process communication protocols and communication data, and the central processor is mainly used to control the entire terminal device, execute software programs, and process the data of software programs. Figure 10 The processor 1011 therein may integrate the functions of the baseband processor and the central processor. Those skilled in the art can understand that the baseband processor and the central processor may also be independent processors, interconnected through technologies such as a bus. Those skilled in the art can understand that the terminal device may include multiple baseband processors to adapt to different network systems, and the terminal device may include multiple central processors to enhance its processing ability. Each component of the terminal device may be connected through various buses. The baseband processor may also be referred to as a baseband processing circuit or a baseband processing chip. The central processor may also be referred to as a central processing circuit or a central processing chip. The function of processing communication protocols and communication data may be built into the processor or stored in the memory in the form of a software program, and the processor executes the software program to implement the baseband processing function.

[0347] The memory is mainly used to store software programs and data. The memory 1012 may exist independently and be connected to the processor 1011. Optionally, the memory 1012 may be integrated with the processor 1011, for example, integrated within a single chip. Among them, the memory 1012 can store the program code for implementing the technical solutions of the embodiments of the present application and is controlled by the processor 1011 to execute. The various computer program codes being executed may also be regarded as the driver programs of the processor 1011.

[0348] Figure 10 Only one memory and one processor are shown. In an actual terminal device, there may be multiple processors and multiple memories. The memory may also be referred to as a storage medium or a storage device, etc. The memory may be a storage element on the same chip as the processor, that is, an on-chip storage element, or an independent storage element. The embodiments of the present application do not make any limitations in this regard.

[0349] The transceiver 1013 can be used to support the reception or transmission of radio frequency signals between a communication device and a terminal. The transceiver 1013 can be connected to the antenna 1015. The transceiver 1013 includes a transmitter Tx and a receiver Rx. Specifically, one or more antennas 1015 can receive radio frequency signals. The receiver Rx of the transceiver 1013 is used to receive the radio frequency signals from the antenna, convert the radio frequency signals into digital baseband signals or digital intermediate frequency signals, and provide the digital baseband signals or digital intermediate frequency signals to the processor 1011 so that the processor 1011 can perform further processing on the digital baseband signals or digital intermediate frequency signals, such as demodulation processing and decoding processing. In addition, the transmitter Tx in the transceiver 1013 is also used to receive the modulated digital baseband signals or digital intermediate frequency signals from the processor 1011, convert the modulated digital baseband signals or digital intermediate frequency signals into radio frequency signals, and transmit the radio frequency signals through one or more antennas 1015. Specifically, the receiver Rx can selectively perform one-stage or multi-stage down-conversion processing and analog-to-digital conversion processing on the radio frequency signals to obtain digital baseband signals or digital intermediate frequency signals, and the order of the down-conversion processing and the analog-to-digital conversion processing can be adjusted. The transmitter Tx can selectively perform one-stage or multi-stage up-conversion processing and digital-to-analog conversion processing on the modulated digital baseband signals or digital intermediate frequency signals to obtain radio frequency signals, and the order of the up-conversion processing and the digital-to-analog conversion processing can be adjusted. Digital baseband signals and digital intermediate frequency signals can be collectively referred to as digital signals.

[0350] The transceiver 1013 can also be referred to as a transceiver unit, a transceiver, a transceiver device, etc. Optionally, the devices used to implement the receiving function in the transceiver unit can be regarded as the receiving unit, and the devices used to implement the sending function in the transceiver unit can be regarded as the sending unit, that is, the transceiver unit includes a receiving unit and a sending unit. The receiving unit can also be referred to as a receiver, an input port, a receiving circuit, etc., and the sending unit can be referred to as a transmitter, a transmitter, or a transmitting circuit, etc.

[0351] It should be noted that Figure 10 The illustrated communication device 1000 can specifically be used to implement the steps implemented by the network device in the foregoing method embodiments and achieve the corresponding technical effects of the network device. Figure 10 For the specific implementation manners of the illustrated communication device 1000, reference can be made to the descriptions in the foregoing method embodiments, and details are not described herein again.

[0352] Please refer to Figure 11 , which is a schematic structural diagram of the communication device involved in the foregoing embodiments provided by the embodiments of the present application.

[0353] It can be understood that the communication device 110 includes, for example, modules, units, components, circuits, or interfaces, etc., which are appropriately configured together to execute the technical solutions provided in this application. The communication device 110 can be the terminal device or network device described above, or a component (such as a chip) in these devices, for implementing the methods described in the following method embodiments. The communication device 110 includes one or more processors 111. The processor 111 can be a general-purpose processor or a dedicated processor, etc. For example, it can be a baseband processor or a central processing unit. The baseband processor can be used to process communication protocols and communication data, and the central processing unit can be used to control the communication device (such as, a RAN node, a terminal, or a chip, etc.), execute software programs, and process the data of software programs.

[0354] Optionally, in one design, the processor 111 can include a program 113 (sometimes also referred to as code or instructions), and the program 113 can be run on the processor 111, so that the communication device 110 executes the methods described in the following embodiments. In another possible design, the communication device 110 includes a circuit ( Figure 11 not shown).

[0355] Optionally, the communication device 110 can include one or more memories 112, on which there is a program 114 (sometimes also referred to as code or instructions), and the program 114 can be run on the processor 111, so that the communication device 110 executes the methods described in the above method embodiments.

[0356] Optionally, the processor 111 and / or the memory 112 can include AI modules 117, 118, and the AI modules are used to implement AI-related functions. The AI modules can be implemented in a software, hardware, or a combination of software and hardware manner. For example, the AI module 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.

[0357] Optionally, data can also be stored in the processor 111 and / or the memory 112. The processor and the memory can be provided separately or integrated together.

[0358] Optionally, the communication device 110 can further include a transceiver 115 and / or an antenna 116. The processor 111 is sometimes also referred to as a processing unit, which controls the communication device (such as a RAN node or a terminal). The transceiver 115 is sometimes also referred to as a transceiver unit, a transceiver, a transceiver circuit, or a transceiver, etc., and is used to implement the transceiver function of the communication device through the antenna 116.

[0359] Among them, Figure 7 the processing unit 701 shown may be the processor 111. Figure 7 The transceiver unit 702 shown may be a communication interface, and this communication interface may be Figure 11 the transceiver 115 in, and this transceiver 115 may include an input interface and an output interface. Alternatively, this transceiver 115 may also be a transceiver circuit, and this transceiver circuit may include an input interface circuit and an output interface circuit.

[0360] The embodiment of the present application also provides a computer-readable storage medium, which is used to store one or more computer-executable instructions. When the computer-executable instructions are executed by a processor, the processor executes the method described in the possible implementation manners of the first node or the control node in the foregoing embodiments.

[0361] The embodiment of the present application also provides a computer program product (or computer program). When the computer program product is executed by the processor, the processor executes the method of the possible implementation manners of the foregoing first node or control node.

[0362] The embodiment of the present application also provides a chip system, which includes at least one processor and is used to support the communication device to implement the functions involved in the possible implementation manners of the foregoing communication device. Optionally, the chip system further includes an interface circuit, and the interface circuit provides program instructions and / or data for the at least one processor. In a possible design, the chip system may further include a memory, which is used to store the necessary program instructions and data of the communication device. The chip system may be composed of chips or may include chips and other discrete devices, where the communication device may specifically be the first node or the control node in the foregoing method embodiments.

[0363] The embodiment of the present application also provides a communication system, and the network system architecture includes the first node and the control node in any of the foregoing embodiments.

[0364] Optionally, the communication system further includes other nodes among the N distributed nodes, such as the second node.

[0365] Optionally, the communication system further includes a data receiving node.

[0366] Optionally, the communication system further includes a central node.

[0367] In several embodiments provided in the present application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces, and the indirect couplings or communication connections of devices or units can be in electrical, mechanical, or other forms.

[0368] The units described as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0369] In addition, each functional unit in various embodiments of the present application can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of a software functional unit. If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present application, in essence, or the part that makes a contribution, or all or part of the technical solution, can be embodied in the form of a software product. This 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 various embodiments of the present application. The foregoing storage medium includes: USB flash drives, mobile hard disks, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical discs, and other various media that can store program codes.

Claims

1. A communication method, characterized in that, it includes: receiving first configuration information for configuring AI data collection; sending first AI data, where the first AI data is collected based on the first configuration information; wherein, the first AI data is used for model processing of a first AI model.

2. The method according to claim 1, characterized in that, the method is applied to a first node, and the first node is one of N distributed nodes, where N is an integer greater than 1; the receiving of the first configuration information includes: receiving the first configuration information from a control node, where the control node is used to control the data collection of the N distributed nodes; or, receiving the first configuration information from the control node through a central node; or, receiving the first configuration information from a second node, where the second node is a node different from the first node among the N distributed nodes.

3. The method according to claim 1 or 2, characterized in that, the first configuration information is the configuration information corresponding to the first node among M configuration information, and the M configuration information is at least used to configure the AI data collection of M distributed nodes among the N distributed nodes, where M is less than or equal to N; the method further includes: sending one or more of the M configuration information to at least one node among the M distributed nodes.

4. The method according to any one of claims 1 to 3, characterized in that, the method is applied to a first node, and the first node is one of N distributed nodes, where N is an integer greater than or equal to 1; the sending of the first AI data includes: sending the first AI data to a data receiving node, where the data receiving node is used for AI data collection; or, sending the first AI data to a central node.

5. The method according to claim 4, characterized in that, the method further includes: sending the data of the first node to other nodes among the N distributed nodes, and the first AI data is used for the data collection of the other nodes.

6. The method according to any one of claims 1 to 5, characterized in that, the method further includes: receiving data from other nodes among the N distributed nodes, and the data of the other nodes is used to determine part or all of the first AI data.

7. A communication method, characterized in that, it includes: determining first configuration information for AI data collection; sending the first configuration information.

8. The method according to claim 7, characterized in that, the method is applied to a control node, where the control node is used to control the data collection of N distributed nodes, and N is an integer greater than or equal to 1; the first configuration information is used for the AI data collection of a first AI node among the N distributed nodes; the sending of the first configuration information includes: sending the first configuration information to the first node; or, sending the first configuration information to the first node through a central node; or, Send the first configuration information to the first node through a second node, where the second node is a node different from the first node among the N distributed nodes.

9. The method according to claim 8, wherein, the first configuration information is one of the K configuration information sent, and the K configuration information is respectively used to configure the AI data collection of the N distributed nodes, and K is less than or equal to N.

10. The method according to any one of claims 7 to 9, wherein, the method further includes: Receiving first AI data, where the first AI data is collected based on the first configuration information.

11. The method according to claim 10, wherein, the method is applied to a control node, and the control node is used to control the data collection of N distributed nodes, where N is an integer greater than or equal to 1; the first configuration information is used for the AI data collection of the first AI node among the N distributed nodes; the receiving the first AI data includes: Receiving the first AI data from the first node; Receiving the first AI data through a central node.

12. The method according to any one of claims 1 to 11, wherein, the first configuration information includes at least one of the following: The identifier of the AI model corresponding to the collected AI data; The identifier of the AI function corresponding to the collected AI data; The identifier of the source node of the collected AI data; The indication information indicating the model processing corresponding to the collected AI data; The indication information indicating that the configuration method of the configuration information for AI data collection is centralized or decentralized; The indication information indicating that the AI data collection method is centralized or decentralized; The indication information indicating the AI data characteristics of the collected AI data; The indication information indicating the AI data processing of the collected AI data; The indication information indicating the collection period of the AI data; The indication information indicating the transmission information of the collected AI data.

13. The method according to any one of claims 1 to 12, wherein, the model processing includes at least one of model training, model inference, and model monitoring.

14. The method according to any one of claims 1 to 13, wherein, the first AI data includes at least one of the AI data used in the model training stage of the first AI model, the AI data used in the model inference stage of the first AI model, and the AI data used in the model monitoring stage of the first AI model.

15. The method according to claim 14, wherein, the AI data used in the model training stage of the first AI model includes at least one of the input data, feature data, and label data for the training of the first AI model; the AI data used in the model inference stage of the first AI model includes at least one of the input data, feature data, and inference result data for the inference of the first AI model; The AI data used in the model monitoring phase of the first AI model includes at least one of input data, feature data, label data, inference result data, AI model performance data, and communication performance data for monitoring the first AI model.

16. A communication device, characterized in that, it includes a module for executing the method according to any one of claims 1 to 15.

17. A communication device, characterized in that, it includes at least one processor, and the at least one processor is coupled to a memory; the at least one processor is used for executing the method according to any one of claims 1 to 15.

18. The communication device according to claim 17, characterized in that, the communication device is a chip or a chip system.

19. A readable storage medium, characterized in that, the storage medium stores a computer program or instruction, and when the computer program or instruction is executed by a communication device, the method according to any one of claims 1 to 15 is implemented.

Images