Communication method and communication apparatus

Abstract

Provided in the present application are a communication method and a communication apparatus. The method comprises: receiving a second dataset and a first identifier, wherein the first identifier is the same as an identifier of a first dataset, or is associated with the identifier of the first dataset; and on the basis of at least one dataset, performing model development on a first model or an initial model, so as to obtain a second model, wherein the at least one dataset comprises the second dataset, or comprises the first dataset and the second dataset, the first model is determined on the basis of the first dataset and the initial model, and the model development comprises model training, offline engineering or model enhancement. On the basis of the present application, if a first apparatus needs to send to a second apparatus a second dataset supplementary to a first dataset, the first apparatus may use an identifier of the first dataset as an identifier of the second dataset, or may associate the identifier of the first dataset with the identifier of the second dataset, thereby not only facilitating a reduction in air interface overheads, but also preventing the second apparatus from performing repeated training.

Description

Communication methods and communication devices

[0001] This application claims priority to Chinese Patent Application No. 202411988876.1, filed with the China National Intellectual Property Administration on December 30, 2024, entitled "Communication Method and Communication Device", the entire contents of which are incorporated herein by reference. Technical Field

[0002] This application relates to the field of wireless communication, and more particularly to a communication method and a communication device. Background Technology

[0003] Currently, artificial intelligence (AI) has been introduced into wireless communication networks and has been widely applied in many application scenarios of air interface technology, such as AI-based channel state information (CSI) prediction, AI-based beam management, and AI-based CSI feedback, playing an increasingly important role.

[0004] In practical applications of AI models, the devices acquiring and / or managing datasets may differ from those training the AI ​​model. Therefore, the devices acquiring and / or managing datasets need to send the datasets to the devices training the AI ​​model, enabling the latter to develop the model based on the received datasets. For example, the network side can provide datasets to the terminal side for model development and deployment. Thus, how to transmit datasets while minimizing air interface overhead has become a pressing technical problem. Summary of the Invention

[0005] This application provides a communication method and a communication device to reduce the overhead of transmitting data sets over the air interface.

[0006] In a first aspect, a communication method is provided, the method comprising: receiving a second dataset and a first identifier; wherein the first identifier is the same as, or associated with, the identifier of the first dataset, and a second model is obtained by model development on a first model or an initial model based on at least one dataset; wherein at least one dataset includes the second dataset, or includes the first dataset and the second dataset; the first model is determined based on the first dataset and the initial model, and the model development includes one or more of the following: model training, offline engineering, or model enhancement.

[0007] This method can be executed by a second device, which can be replaced by a device on the terminal device side, a device on the network device side, or a device on the core network element side.

[0008] The equipment on the terminal device side can include the terminal device itself, the communication module within the terminal device, or the circuits or chips within the terminal device responsible for communication functions (such as a modem chip, also known as a baseband chip, or a system-on-chip (SoC) chip containing a modem core, or a system-in-package (SIP) chip, etc.). Alternatively, the equipment on the terminal device side can include AI entities on the terminal device side. AI entities on the terminal device side can be the terminal device itself, or AI entities serving the terminal device, such as servers, such as over-the-top (OTT) servers or cloud servers.

[0009] Network-side devices can include the network device itself, communication modules within the network device, or circuits or chips responsible for communication functions within the network device (such as modem chips, also known as baseband chips, or system-on-a-chip (SoC) chips or SIP chips containing modem cores, etc.). Alternatively, network-side devices can include AI entities on the network device side. These AI entities can be the network device itself or AI entities serving the network device, such as radio access network (RAN) intelligent controllers (RICs), operation administration and maintenance (OAM) systems, or servers, such as OTT servers or cloud servers.

[0010] The equipment on the core network element side can include the core network element itself, the communication module within the core network element, or the circuits or chips (such as modem chips, also known as baseband chips, or system-on-a-chip (SoC) chips or SIP chips containing modem cores) responsible for communication functions within the core network element. Alternatively, the equipment on the core network element side can include AI entities on the core network element side. These AI entities can be the core network element itself or AI entities serving the core network element, such as servers, like OTT servers or cloud servers.

[0011] Based on the above technical solution, if the dataset sender (e.g., the first device) needs to send a supplementary second dataset to the second device based on the first dataset, so that the second device can use the second dataset to develop a first model or initial model associated with the first dataset, the first device can use the identifier of the first dataset as the identifier of the second dataset, or associate the identifier of the first dataset with the identifier of the second dataset. This helps reduce overhead and avoids redundant training by the second device. For example, since the identifier of the first dataset is the same as or associated with the identifier of the second dataset, the first device does not need to send both the first and second datasets to the second device simultaneously. Instead, by sending the second dataset to the second device, the second device can use both the second dataset and the previously received first dataset for model development. As another example, since the identifier of the first dataset is the same as or associated with the identifier of the second dataset, by sending the second dataset to the second device, the second device can use the second dataset to develop the first model, thus avoiding redundant training using the first dataset.

[0012] The data set sending end (e.g., the first device) can be a device on the terminal device side, a device on the network device side, or a device on the core network element side.

[0013] For example, developing a second model based on at least one dataset to obtain a first model or an initial model includes: a second device performing initial model training or offline engineering on the initial model based on the first dataset and the second dataset to obtain the second model.

[0014] For example, developing a second model based on at least one dataset to obtain a first model or an initial model includes: a second device performing initial model training or offline engineering on the initial model based on the second dataset to obtain the second model.

[0015] For example, developing a second model based on at least one dataset to obtain a first model or an initial model includes: a second device performing offline engineering, model retraining, model fine-tuning, model updating, or model enhancement on the first model based on the first dataset and the second dataset to obtain a second model.

[0016] For example, developing a second model based on at least one dataset to obtain a first model or an initial model includes: a second device performing offline engineering, model retraining, model fine-tuning, model updating, or model enhancement on the first model based on the second dataset to obtain the second model.

[0017] In conjunction with the first aspect, in some implementations of the first aspect, the method further includes: sending an identifier of at least one dataset, wherein the identifier of the at least one dataset includes the identifier of the first dataset.

[0018] Based on the above technical solution, the second device can send at least one identifier of a dataset to the first device, so that the first device can use the identifier of the first dataset as the first identifier, or associate the identifier of the first dataset with the first identifier.

[0019] In conjunction with the first aspect, in some implementations of the first aspect, the first identifier is associated with the identifier of the first dataset, and the method further includes: receiving first information, the first information indicating the association relationship of at least two identifiers, the association relationship of the at least two identifiers including the association relationship between the first identifier and the identifier of the first dataset.

[0020] Based on the above technical solution, when the second device receives the first information, it can determine the association between the first identifier and the identifier of the first dataset based on the first information, which is beneficial for the second device to determine the second dataset for model development of the first model or the initial model based on the association between the first identifier and the identifier of the first dataset.

[0021] In conjunction with the first aspect, in some implementations of the first aspect, the first dataset corresponds to the first performance index. The method further includes: determining that the second dataset corresponds to the first performance index; and developing a second model based on at least one dataset to obtain a first model or an initial model, including: developing a second model based on at least one dataset and the first performance index to obtain a first model or an initial model, wherein the second model satisfies the first performance index.

[0022] Based on the above technical solution, if the dataset sending end (e.g., the first device) needs to send a supplementary second dataset to the second device based on the first dataset, then, assuming the first and second datasets correspond to the same performance metrics, the second device can avoid using incorrect performance metrics to monitor the model development process based on the second dataset, thereby affecting model performance. Furthermore, if the dataset sending end sends a supplementary second dataset, then, assuming the first and second datasets correspond to the same performance metrics, the second device can reuse the first performance metrics corresponding to the first dataset to monitor the model development process based on the second dataset, thereby reducing the signaling overhead caused by repeatedly indicating the performance metrics corresponding to the second dataset.

[0023] Optionally, if the first identifier is the same as the identifier of the first dataset, then the second device determines that the second dataset corresponds to the first performance metric. In other words, if the first identifier is the same as the identifier of the first dataset, then the second device determines that the first dataset and the second dataset correspond to the same performance metric.

[0024] In conjunction with the first aspect, in some implementations of the first aspect, the first dataset corresponds to the first performance metric. The method further includes: receiving a second performance metric, the second performance metric corresponding to the first dataset and different from the first performance metric; determining that the second dataset corresponds to the second performance metric; and developing a second model based on at least one dataset to obtain a first model or an initial model, including: developing a second model based on at least one dataset and a second performance metric to obtain a second model, wherein the second model satisfies the second performance metric.

[0025] Based on the above technical solution, if the dataset sending end (e.g., the first device) needs to send a supplementary second dataset to the second device based on the first dataset, then by defining the same performance metrics for the first dataset and the second dataset, it is possible to avoid the second device using incorrect performance metrics to monitor the model development process based on the second dataset, thereby affecting the model's performance.

[0026] Optionally, if the first identifier is the same as the identifier of the first dataset, the second device determines that the second dataset corresponds to the second performance metric.

[0027] In conjunction with the first aspect, in some implementations of the first aspect, the first dataset corresponds to the first performance metric. The method further includes: receiving a third performance metric, which corresponds to a second dataset; determining that the first dataset corresponds to the first performance metric; and developing a second model based on at least one dataset to obtain a first model or an initial model, including: developing a second model based on at least one dataset and the third performance metric to obtain a first model or an initial model, wherein the second model satisfies the third performance metric.

[0028] Based on the above technical solution, if the dataset sending end (e.g., the first device) needs to send a supplementary second dataset to the second device based on the first dataset, then by defining the same performance metrics for the first dataset and the second dataset, it is possible to avoid the second device using incorrect performance metrics to monitor the model development process based on the second dataset, thereby affecting the model's performance.

[0029] Optionally, if the first identifier is the same as the identifier of the first dataset, the second device determines that the first dataset corresponds to the third performance metric.

[0030] In conjunction with the first aspect, in some implementations of the first aspect, the first dataset corresponds to the first performance metric. The method further includes: receiving a second performance metric, which corresponds to the first dataset and is different from the first performance metric; receiving a third performance metric, which corresponds to the second dataset; and developing a second model based on at least one dataset to obtain a first model or an initial model, including: developing a second model based on at least one dataset and at least one performance metric to obtain a first model or an initial model, wherein the second model satisfies at least one performance metric; wherein at least one dataset includes the second dataset, and at least one performance metric includes the second performance metric; or, at least one dataset includes the first dataset and the second dataset, and at least one performance metric includes both the second and third performance metrics.

[0031] Based on the above technical solution, if the dataset sending end (e.g., the first device) needs to send a supplementary second dataset to the second device based on the first dataset, it can also send the performance indicators corresponding to the first dataset and the second dataset to the second device respectively. This can prevent the second device from using incorrect performance indicators to monitor the model development process based on the second dataset, thereby affecting the performance of the model.

[0032] Secondly, a communication method is provided, the method comprising: sending a second dataset and a first identifier, the first identifier being the same as or associated with the identifier of the first dataset; the first dataset being used for model development of a first model or an initial model, the model development including one or more of the following: model training or model enhancement, the first model being determined based on the first dataset and the initial model.

[0033] This method can be executed by a first device, which can be replaced by a device on the terminal device side, a device on the network device side, or a device on the core network element side. Further description of the device on the terminal device side, the device on the network device side, or the device on the core network element side can be found in the first aspect described above.

[0034] In conjunction with the second aspect, in some implementations of the second aspect, the method further includes: receiving an identifier of at least one dataset, wherein the identifier of the at least one dataset includes an identifier of the first dataset.

[0035] In conjunction with the second aspect, in some implementations of the second aspect, the first identifier is associated with the identifier of the first dataset, and the method further includes: sending first information, the first information indicating the association relationship of at least two identifiers, the association relationship of the at least two identifiers including the association relationship between the first identifier and the identifier of the first dataset.

[0036] In conjunction with the second aspect, in some implementations of the second aspect, the first dataset corresponds to the first performance metric. The method further includes: sending a second performance metric, which corresponds to the first dataset and is different from the first performance metric.

[0037] In conjunction with the second aspect, in some implementations of the second aspect, the first dataset corresponds to the first performance metric. The method further includes sending a third performance metric, which corresponds to the second dataset.

[0038] It should be understood that the methods provided in the second aspect correspond to those in the first aspect. The descriptions and technical effects of the various implementation methods in the second aspect can be found in the relevant descriptions in the first aspect, and will not be repeated here.

[0039] Thirdly, an apparatus is provided. This apparatus may include functional modules corresponding to each of the methods / operations / steps / actions described in any possible implementation of the first aspect, or may include functional modules corresponding to each of the methods / operations / steps / actions described in any of the second aspects. The module may be a hardware circuit, software, or a combination of hardware circuitry and software implementation.

[0040] In one design, the device may include a processing module and a communication module. The communication module is used to perform the sending and receiving actions performed by the second device in the method described in the first aspect above, while the processing module is used to perform processing-related actions performed by the second device in the method described in the first aspect above.

[0041] In one design, the device can be a terminal device, or a device, module, circuit, or chip configured in the terminal device, or a device that can be used in conjunction with the terminal device, such as an OTT host or cloud server.

[0042] In one design, the device may include a processing module and a communication module. The communication module is used to perform the sending and receiving actions performed by the first device in the method described in the second aspect above, while the processing module is used to perform processing-related actions performed by the first device in the method described in the second aspect above.

[0043] In one design, the device can be a network device, or a device, module, circuit, or chip configured in the network device, or a device that can be used in conjunction with the network device, such as an intelligent network element with a radio access network (RAN) intelligent controller (RIC) deployed thereon.

[0044] Fourthly, an apparatus is provided, comprising a processor and a storage medium storing instructions that, when executed by the processor, cause a method as described in the first aspect or any possible implementation thereof to be implemented, or cause a method as described in the second aspect or any possible implementation thereof to be implemented.

[0045] Fifthly, an apparatus is provided, comprising a processing circuit for processing data and / or information such that a method as in the first aspect or any possible implementation thereof is implemented, or a method as in the second aspect or any possible implementation thereof is implemented.

[0046] The processing circuit may include one or more processors, or all or part of the circuitry in one or more processors used for control or processing functions.

[0047] Optionally, the apparatus may further include a memory for storing programs or instructions, and the processor for running the programs or instructions to implement the methods as described in the first aspect or any possible implementation thereof, or to implement the methods as described in the second aspect or any possible implementation thereof.

[0048] Optionally, the device may also include the transceiver circuit, or an input / output interface.

[0049] In a sixth aspect, a chip is provided, including processing circuitry for running a program or instructions to cause the method as described in the first aspect or any possible implementation thereof to be implemented, or to cause the method as described in the second aspect or any possible implementation thereof to be implemented.

[0050] Optionally, the chip may further include a memory for storing programs or instructions.

[0051] Optionally, the chip may also include transceiver circuitry, or input / output interfaces.

[0052] A seventh aspect provides a computer-readable storage medium comprising instructions that, when executed by a processor, cause the method as described in the first aspect or any possible implementation thereof to be implemented, or cause the method as described in the second aspect or any possible implementation thereof to be implemented.

[0053] Eighthly, a computer program product is provided, the computer program product comprising computer program code or instructions, which, when executed, cause the method as described in the first aspect and any possible implementation thereof to be implemented, or cause the method as described in the second aspect and any possible implementation thereof to be implemented.

[0054] Ninth aspect, a communication system is provided, the communication system including means for performing the first aspect and any possible implementation thereof, or including means for performing the second aspect and any possible implementation thereof.

[0055] It should be understood that the third to ninth aspects of this application correspond to the technical solutions of the first to second aspects of this application, and the beneficial effects obtained by each aspect and the corresponding feasible implementation are similar, and will not be repeated here. Attached Figure Description

[0056] Figure 1 is a schematic diagram of a communication system applicable to the communication method of this application embodiment;

[0057] Figure 2 is a schematic diagram of another communication system applicable to the communication method of this application embodiment;

[0058] Figure 3 is a schematic diagram of a possible application framework in a communication system;

[0059] Figure 4 is a schematic diagram of another possible application framework in a communication system;

[0060] Figure 5 is a schematic diagram of CSI feedback using an auto-encoder (AE) model provided in an embodiment of this application;

[0061] Figure 6 is a schematic diagram of a deep neural network (DNN);

[0062] Figure 7 shows an example of a neuron structure;

[0063] Figure 8 is a schematic diagram of the data set docking between the network side and the terminal side provided in an embodiment of this application;

[0064] Figure 9 is a schematic diagram of model interoperability between the network side and the terminal side provided in an embodiment of this application;

[0065] Figure 10 is a schematic flowchart of the communication method provided in an embodiment of this application;

[0066] Figure 11 is a schematic flowchart of the communication method provided in an embodiment of this application;

[0067] Figure 12 is a schematic flowchart of the communication method provided in an embodiment of this application;

[0068] Figure 13 is a schematic block diagram of a communication device provided in an embodiment of this application;

[0069] Figure 14 is another schematic block diagram of the communication device provided in an embodiment of this application;

[0070] Figure 15 is a schematic block diagram of the AI ​​processor provided in an embodiment of this application. Detailed Implementation

[0071] The technical solutions in this application will now be described with reference to the accompanying drawings.

[0072] To facilitate understanding of the embodiments of this application, the following points will be explained first:

[0073] First, in this application, the terminal side can also be referred to as the UE (user equipment) side, including: terminal equipment, components deployed in the terminal equipment (such as circuits or chips inside the terminal equipment), equipment deployed outside the terminal equipment (such as OTT hosts or cloud servers), or components deployed in equipment outside the terminal equipment (such as circuits or chips inside the equipment). The network (NW) side includes: network equipment communicating with the terminal equipment, components deployed in the network equipment (such as circuits or chips with near real-time RAN intelligent control functions inside the network equipment), equipment deployed outside the network equipment (such as intelligent network elements, for example, intelligent network elements with near real-time RAN intelligent control functions), or components deployed in the intelligent network element (such as circuits or chips inside the intelligent network element). The network equipment may include: access network equipment, core network equipment, or operation administration and maintenance (OAM) equipment.

[0074] Second, in this application, the indication includes direct indication (also known as explicit indication) and indirect indication (also known as implicit indication). Directly indicating information A means including information A; indirectly indicating information A can mean indicating information A through the correspondence between information A and information B and by directly indicating information B; or by indicating information A through a preset rule that can be used to determine A based on B and by directly indicating information B. The correspondence between information A and information B, and the preset rule, can be predefined, pre-stored, pre-burned, or pre-configured.

[0075] Third, in this application, "at least one" means one or more, and "more than one" means two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can mean: A alone, A and B simultaneously, or B alone, where A and B can be singular or plural. The character " / " generally indicates an "or" relationship between the preceding and following related objects, but it does not exclude the possibility of indicating an "and" relationship; the specific meaning can be understood in context. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one of a, b, or c can mean: a, b, c; a and b; a and c; b and c; or a and b and c. Here, a, b, and c can be single or multiple.

[0076] Fourth, the use of prefixes such as "first" and "second" in this application is merely for the purpose of distinguishing and describing different things belonging to the same category, and does not constrain the order, size, or quantity of things. For example, "first information" and "second information" are simply different information, and do not limit the quantity of information, the order of transmission, or the relationship of priority.

[0077] Fifth, in this application, "send" and "receive" indicate the direction of signal transmission. For example, "send information to XX" can be understood as the destination of the information being XX, which can include direct transmission via the air interface or indirect transmission by other units or modules via the air interface. "Receive information from YY" can be understood as the source of the information being YY, which can include direct reception from YY via the air interface or indirect reception from YY by other units or modules via the air interface. "Send" can also be understood as the "output" of the chip interface, and "receive" can also be understood as the "input" of the chip interface. In other words, sending and receiving can occur between devices, such as between a terminal device and a computing node, or within a device, such as between components, modules, chips, software modules, or hardware modules within the device via a bus, wiring, or interface.

[0078] Sixth, in the embodiments of this application, "when," "if," and "if" all refer to the device making corresponding processing under certain objective circumstances, and are not limited to a time, nor do they require the device to make a judgment action when it is implemented, nor do they mean that there are other limitations.

[0079] Seventh, in this application, the words "example," "exemplarily," "for example," or "such as" are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "example," "exemplarily," "for example," or "such as" in this application should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the words "example," "exemplarily," "for example," or "such as" is intended to present the relevant concepts in a specific manner.

[0080] The technical solutions provided in this application can be applied to various communication systems, such as: 5th generation (5G) or new radio (NR) systems, long term evolution (LTE) systems, LTE frequency division duplex (FDD) systems, LTE time division duplex (TDD) systems, wireless local area network (WLAN) systems, satellite communication systems, future communication systems, or integrated systems of multiple systems. The technical solutions provided in this application can also be applied to device-to-device (D2D) communication, vehicle-to-everything (V2X) communication, machine-to-machine (M2M) communication, machine-type communication (MTC), and Internet of Things (IoT) communication systems or other communication systems.

[0081] In a communication system, one network element can send signals to or receive signals from another network element. These signals can include information, signaling, or data. The term "network element" can also be replaced by an entity, network entity, device, communication equipment, communication module, node, communication node, etc. This disclosure uses a network element as an example. For instance, a communication system can include at least one terminal device and at least one network device. The network device can send downlink signals to the terminal device, and / or the terminal device can send uplink signals to the network device. It is understood that the terminal device in this disclosure can be replaced by a first network element, and the network device can be replaced by a second network element, both performing the corresponding communication methods described in this disclosure.

[0082] Figure 1 is a schematic diagram of a communication system applicable to the communication method of this application embodiment. As shown in Figure 1, the communication system 100A may include at least one access network device, such as access network device 110 shown in Figure 1; the communication system 100A may also include at least one terminal device, such as terminal device 120 and terminal device 130 shown in Figure 1. Access network device 110 and terminal devices (such as terminal device 120 and terminal device 130) can communicate via a wireless link. The communication devices in this communication system, for example, access network device 110 and terminal device 120, can communicate via multi-antenna technology.

[0083] In wireless communication networks, such as mobile communication networks, the services supported by the networks are becoming increasingly diverse, thus requiring increasingly diverse demands. For example, networks need to support ultra-high speeds, ultra-low latency, and / or massive connectivity. This characteristic makes network planning, network configuration, and / or resource scheduling increasingly complex. Furthermore, as network functions become more powerful, such as supporting higher spectrum levels, supporting higher-order multiple-input multiple-output (MIMO) technologies, supporting beamforming (BF), and supporting beam management, network energy efficiency has become a hot research topic. These new demands, new scenarios, and new characteristics bring unprecedented challenges to network planning, operation, and efficient operation. To meet these challenges, artificial intelligence (AI) technology can be introduced into wireless communication networks to achieve network intelligence. To support AI technology in wireless networks, AI nodes may also be introduced.

[0084] Figure 2 is a schematic diagram of another communication system applicable to the communication method of this application embodiment. Compared with the communication system 100A shown in Figure 1, the communication system 100B shown in Figure 2 further includes an AI network element 140. The AI ​​network element 140 is used to perform AI-related operations, such as building training datasets or training AI models. The AI ​​network element can also be simply referred to as an intelligent network element.

[0085] In one possible implementation, access network device 110 can send data related to the training of the AI ​​model to AI network element 140, which then constructs a training dataset and trains the AI ​​model. For example, the data related to the training of the AI ​​model may include data reported by terminal devices. AI network element 140 can send the results of operations related to the AI ​​model to access network device 110, which then forwards them to the terminal devices. For example, the results of operations related to the AI ​​model may include at least one of the following: a trained AI model, model evaluation results, or test results. Exemplarily, a portion of the trained AI model may be deployed on access network device 110, and another portion on terminal devices 120 and / or 130. Alternatively, the trained AI model may be deployed on access network device 110. Or, the trained AI model may be deployed on terminal devices 120 and / or 130.

[0086] It should be understood that Figure 2 is only used as an example of the AI ​​network element 140 being directly connected to the access network device 110. In other scenarios, the AI ​​network element 140 can also be connected to the terminal device. Alternatively, the AI ​​network element 140 can be connected to both the access network device 110 and the terminal device simultaneously. Alternatively, the AI ​​network element 140 can also be connected to the access network device 110 through a third-party network element. This application embodiment does not limit the connection relationship between the AI ​​network element and other network elements. For example, the AI ​​network element 140 can also be set as a module in the access network device and / or the terminal device, for example, in the access network device 110 or the terminal device shown in Figure 1.

[0087] It should be noted that Figures 1 and 2 are simplified schematic diagrams for ease of understanding. For example, the communication system may also include other devices, such as wireless relay devices and / or wireless backhaul devices, which are not shown in Figures 1 and 2. In practical applications, the communication system may include multiple access network devices and multiple terminal devices. The embodiments of this application do not limit the number of access network devices and terminal devices included in the communication system.

[0088] In the embodiments of this application, the terminal device may also be referred to as UE, access terminal, user unit, user station, mobile station, mobile station, remote station, remote terminal, mobile device, user terminal, terminal, wireless communication device, user agent, or user equipment.

[0089] Terminal devices can be devices that provide voice / data, such as handheld devices with wireless connectivity, in-vehicle devices, etc. Currently, examples of terminals include: mobile phones, tablets, laptops, PDAs, mobile internet devices (MIDs), wearable devices, virtual reality (VR) devices, augmented reality (AR) devices, wireless terminals in industrial control, wireless terminals in self-driving, wireless terminals in remote medical surgery, wireless terminals in smart grids, wireless terminals in transportation safety, wireless terminals in smart cities, wireless terminals in smart homes, cellular phones, cordless phones, session initiation protocol (SIP) phones, wireless local loop (WLL) stations, personal digital assistants (PDAs), handheld devices with wireless communication capabilities, computing devices or other processing devices connected to a wireless modem, wearable devices, terminal devices in 5G networks, or future public land mobile communication networks. Terminal devices in a network (PLMN), etc., are not limited to this in the embodiments of this application.

[0090] By way of example and not limitation, in this embodiment, the terminal device can also be a wearable device. Wearable devices, also known as wearable smart devices, are a general term for devices that utilize wearable technology to intelligently design and develop everyday wearables, such as glasses, gloves, watches, clothing, and shoes. Wearable devices are portable devices that are worn directly on the body or integrated into the user's clothing or accessories. Wearable devices are not merely hardware devices, but also achieve powerful functions through software support, data interaction, and cloud interaction. Broadly speaking, wearable smart devices include those that are feature-rich, large in size, and can achieve complete or partial functions without relying on a smartphone, such as smartwatches or smart glasses, as well as those that focus on a specific type of application function and require the use of other devices such as smartphones, such as various smart bracelets and smart jewelry for vital sign monitoring.

[0091] In this embodiment, the device for implementing the functions of the terminal device can be the terminal device itself, or it can be any device capable of supporting the terminal device in implementing those functions, such as a chip system. This device can be installed in or used in conjunction with the terminal device. In this embodiment, the chip system can be composed of chips or may include chips and other discrete components. This embodiment only uses the terminal device as an example to illustrate the device for implementing the functions of the terminal device, and does not constitute a limitation on the solution of this embodiment.

[0092] The access network device in this application embodiment can be a device used to communicate with terminal devices. This access network device can also be called a network device, such as a base station. In this application embodiment, the access network device can refer to a RAN node (or device) that connects the terminal device to the wireless network. A base station can broadly encompass, or be replaced by, various names such as: NodeB, evolved NodeB (eNB), next-generation NodeB (gNB), relay station, access point, transmitting and receiving point (TRP), transmitting point (TP), master station, auxiliary station, motor slide retainer (MSR) node, home base station, network controller, access node, wireless node, access point (AP), transmission node, transceiver node, baseband unit (BBU), remote radio unit (RRU), active antenna unit (AAU), remote radio head (RRH), central unit (CU), distributed unit (DU), radio unit (RU), positioning node, etc. A base station can be a macro base station, micro base station, relay node, donor node, or similar entities, or combinations thereof. A base station can also refer to a communication module, modem, or chip installed within the aforementioned equipment or apparatus. A base station can also be a mobile switching center, a device that performs base station functions in D2D, V2X, and M2M communications, or a device that performs base station functions in future communication systems. A base station can support networks using the same or different access technologies. Optionally, a RAN node can also be a server, wearable device, vehicle, or in-vehicle equipment. For example, the access network equipment in V2X technology can be a roadside unit (RSU). The embodiments of this application do not limit the specific technology or equipment form used in the access network equipment.

[0093] Base stations can be fixed or mobile. For example, a helicopter or drone can be configured to act as a mobile base station, and one or more cells can move depending on the location of the mobile base station. In other examples, a helicopter or drone can be configured as a device to communicate with another base station.

[0094] In some deployments, the access network equipment mentioned in the embodiments of this application may be a device including a CU, or a DU, or a device including both CU and DU, or a device with a control plane CU node (central unit-control plane (CU-CP)) and a user plane CU node (central unit-user plane (CU-UP)) and a DU node. For example, the access network equipment may include gNB-CU-CP, gNB-CU-UP, and gNB-DU.

[0095] In some deployments, multiple RAN nodes collaborate to assist terminals in achieving wireless access, with different RAN nodes each implementing some of the base station's functions. For example, RAN nodes can be CUs, DUs, CU-CPs, CU-UPs, or RUs. CUs and DUs can be configured separately or included in the same network element, such as a BBU. RUs can be included in radio frequency equipment or radio frequency units, such as RRUs, AAUs, or RRHs.

[0096] RAN nodes can support one or more types of fronthaul interfaces, each corresponding to a DU and RU with different functions. If the fronthaul interface between the DU and RU is a common public radio interface (CPRI), the DU is configured to implement one or more baseband functions, and the RU is configured to implement one or more radio frequency functions. If the fronthaul interface between the DU and RU is another type of interface, relative to CPRI, some downlink and / or uplink baseband functions, such as, for downlink, precoding, digital beamforming (BF), or one or more of inverse fast Fourier transform (IFFT) / cyclic prefix addition (CP), are moved from the DU to the RU; and for uplink, one or more of digital beamforming (BF), or fast Fourier transform (FFT) / cyclic prefix removal (CP), are moved from the DU to the RU. In one possible implementation, the interface can be an enhanced common public radio interface (eCPRI). Under the eCPRI architecture, the segmentation between DU and RU differs, corresponding to different categories (Cat) of eCPRI, such as eCPRI Cat A, B, C, D, E, and F.

[0097] Taking eCPRI Cat A as an example, for downlink transmission, layer mapping is used as the dividing line. DU is configured to implement one or more functions preceding layer mapping (i.e., coding, rate matching, scrambling, modulation, and layer mapping), while other functions following layer mapping (e.g., resource element (RE) mapping, digital beamforming (BF), or one or more functions in inverse fast Fourier transform (IFFT) / adding CP) are moved to RU. For uplink transmission, de-RE mapping is used as the dividing line. DU is configured to implement one or more functions preceding de-mapping (i.e., decoding, rate matching de-matching, descrambling, demodulation, inverse discrete Fourier transform (IDFT), channel equalization, and de-RE mapping), while other functions following de-mapping (e.g., digital BF or FFT / removing CP) are moved to RU. Understandably, the functional descriptions of the DU and RU corresponding to various types of eCPRI can be found in the eCPRI protocol, and will not be elaborated here.

[0098] In one possible design, the processing unit in the BBU used to implement baseband functions is called the baseband high (BBH) unit, and the processing unit in the RRU / AAU / RRH used to implement baseband functions is called the baseband low (BBL) unit.

[0099] In different systems, CU (or CU-CP and CU-UP), DU, or RU may have different names, but those skilled in the art will understand their meaning. For example, in an open RAN (ORAN) architecture, CU can also be called open CU (open-CU, O-CU), DU can also be called open DU (open-DU, O-DU), CU-CP can also be called open CU-CP (open-CU-CP) O-CU-CP, CU-UP can also be called open CU-UP (open-CU-UP, O-CU-UP), and RU can also be called open RU (open-RU, O-RU). Any of the units among CU (or CU-CP, CU-UP), DU, and RU in this application can be implemented through software modules, hardware modules, or a combination of software modules and hardware modules.

[0100] In this embodiment, the apparatus for implementing the functions of a network device can be a network device itself; it can also be an apparatus capable of supporting the network device in implementing those functions, such as a chip system, hardware circuit, software module, or a hardware circuit plus a software module. This apparatus can be installed in the network device or used in conjunction with the network device. In this embodiment, the example of a network device being used to implement the functions of a network device is provided only and does not constitute a limitation on the solutions described in this embodiment.

[0101] Network devices and / or terminal devices can be deployed on land, including indoors or outdoors, handheld or vehicle-mounted; they can also be deployed on water; and they can also be deployed in the air on airplanes, balloons, and satellites. This application does not limit the scenario in which the network devices and terminal devices are located. Furthermore, terminal devices and network devices can be hardware devices, or software functions running on dedicated hardware or general-purpose hardware, such as virtualization functions instantiated on a platform (e.g., a cloud platform), or entities that include dedicated or general-purpose hardware devices and software functions. This application does not limit the specific form of the terminal devices and network devices.

[0102] Optionally, the AI ​​node can be deployed in one or more of the following locations within the communication system: access network equipment, terminal equipment, or core network elements. Alternatively, the AI ​​node can also be deployed independently, for example, in a location other than any of the aforementioned devices, such as in the host or cloud server of an over-the-top (OTT) system. The AI ​​node can communicate with other devices in the communication system, which can be, for example, one or more of the following: access network equipment, terminal equipment, or core network elements.

[0103] It is understood that this application does not limit the number of AI nodes. For example, when there are multiple AI nodes, they can be divided based on function, such as different AI nodes being responsible for different functions.

[0104] It can also be understood that AI nodes can be independent devices, or they can be integrated into the same device to achieve different functions. Alternatively, they can be network elements in hardware devices, software functions running on dedicated hardware, or virtualization functions instantiated on a platform (e.g., a cloud platform). This application does not limit the specific form of the aforementioned AI nodes.

[0105] AI nodes can be AI network elements or AI modules.

[0106] Figure 3 illustrates a possible application framework in a communication system. As shown in Figure 3, network elements in the communication system are connected via interfaces (e.g., next-generation (NG) interfaces, Xn interfaces) or air interfaces. The NG interface is the interface between the radio access network and the 5G core network. The Xn interface is the interface between access network devices, and the air interface is the interface between access network devices and terminal devices. These network element nodes, such as core network devices, access network nodes (RAN nodes), terminals, or one or more devices in the OAM, are equipped with one or more AI modules (only one is shown in Figure 3 for clarity). The access network node can be a single RAN node or can include multiple RAN nodes, for example, including CU and DU. The CU and / or DU can also be equipped with one or more AI modules. Optionally, the CU can be further divided into CU-CP and CU-UP. One or more AI models are configured in the CU-CP and / or CU-UP.

[0107] The AI ​​module is used to implement corresponding AI functions. AI modules deployed in different network elements can be the same or different. Depending on the parameter configuration, the AI ​​module can implement different functions. The AI ​​module model can be configured based on one or more of the following parameters: structural parameters (e.g., at least one of the following: number of neural network layers, neural network width, inter-layer connections, neuron weights, neuron activation function, or bias in the activation function), input parameters (e.g., type and / or dimension of input parameters), or output parameters (e.g., type and / or dimension of output parameters). The bias in the activation function can also be referred to as the neural network bias.

[0108] An AI module can have one or more models. A model can infer an output, which includes one or more parameters. The learning, training, or inference processes of different models can be deployed on different nodes or devices, or they can be deployed on the same node or device.

[0109] Network devices can be network devices equipped with one or more AI modules. These network devices can include one or more devices from the core network equipment, access network nodes (RAN nodes), or operation administration and maintenance (OAM) systems shown in Figure 3. For example, the AI ​​module can be a RAN intelligent controller (RIC) as shown in Figure 4, such as a near-real-time RIC (near-RT RIC) or a non-real-time RIC (non-RT RIC). For instance, a near-real-time RIC is located in a RAN node (e.g., in a CU or DU), while a non-real-time RIC is located in OAM, a cloud server, a core network device, or other access network devices. The RIC can obtain subsets from multiple terminal devices from RAN nodes (e.g., CU, CU-CP, CU-UP, DU, and / or RU), reassemble them into a training dataset, and train based on this dataset. Exemplarily, near-real-time RICs and non-real-time RICs can also be configured as separate network elements, and the access network device can be either a near-real-time RIC or a non-real-time RIC.

[0110] Figure 4 illustrates another possible application framework in a communication system. In addition to access network nodes (CU, DU, and RU are shown in the figure) and terminals, the communication system shown in Figure 4 also includes an RIC (Regulator-Integrated Circuit). For example, the RIC could be the AI ​​module shown in Figure 3, which can be used to implement AI-related functions. The RIC includes near-real-time RICs and non-real-time RICs. Non-real-time RICs primarily process non-real-time information, such as data that is not sensitive to latency, with latency on the order of seconds. Real-time RICs primarily process near-real-time information, such as data that is relatively sensitive to latency, with latency on the order of tens of milliseconds.

[0111] The near real-time RIC is used for model training and inference. For example, it can be used to train an AI model and then use that AI model for inference. The near real-time RIC can obtain network-side and / or terminal-side information from RAN nodes (e.g., CU, CU-CP, CU-UP, DU, and / or RU) and / or terminals. This information can be used as training data or inference data. Optionally, the near real-time RIC can deliver inference results to RAN nodes and / or terminals. Optionally, inference results can be exchanged between CU and DU, and / or between DU and RU. For example, the near real-time RIC delivers the inference result to the DU, and the DU sends it to the RU.

[0112] The non-real-time RIC is also used for model training and inference. For example, it can be used to train an AI model and then use that model for inference. The non-real-time RIC can obtain network-side and / or terminal-side information from RAN nodes (e.g., one or more of CU, CU-CP, CU-UP, DU, or RU) and / or terminals. This information can be used as training data or inference data, and the inference results can be delivered to the RAN nodes and / or terminals. Optionally, inference results can be exchanged between CU and DU, and / or between DU and RU; for example, the non-real-time RIC delivers the inference results to the DU, which then forwards them to the RU.

[0113] The near real-time RIC and non-real-time RIC can also be set up as separate network elements. Optionally, the near real-time RIC and non-real-time RIC can also be part of other devices. For example, the near real-time RIC can be set in the RAN node (e.g., in CU, DU), while the non-real-time RIC can be set in the OAM, cloud server, core network device, or other network device.

[0114] To facilitate understanding of the embodiments of this application, the terms involved in this application will be briefly explained below.

[0115] 1. AI Model: A function model that maps an input of a certain dimension to an output of a certain dimension. Its parameters can be obtained through machine learning (ML). For example, f(x) = ax 2 +b is a quadratic function model, which can be viewed as an AI model. a and b correspond to the parameters of the model and can be obtained through machine learning training.

[0116] 2. Autoencoder (AE) model: This generally refers to a network structure consisting of two AI models, such as an encoder and a decoder. Each model can be an independent AI model. AE models are also called bilateral models, two-end models, collaborative models, etc. The encoder and decoder of an AE are usually trained together and can be used in a coordinated manner.

[0117] For example, CSI feedback can be implemented based on the AI ​​model of AE. Figure 5 is a schematic diagram of CSI feedback using an AE model provided in an embodiment of this application.

[0118] As shown in the figure, the terminal side compresses the target CSI using an encoder, while the network side reconstructs the CSI using a decoder. For example, the terminal side can use the target CSI (i.e., V) as input to the encoder, which compresses the target CSI to obtain the compressed CSI (i.e., C). The terminal side can quantize the compressed CSI to obtain CSI feedback information. The terminal side can then send this CSI feedback information to the network side, for example, through a CSI report.

[0119] The network side can first dequantize the CSI feedback information to obtain compressed CSI with quantization loss (i.e., The network side can use this compressed CSI as input to the decoder, which can then decompress the CSI to obtain the reconstructed CSI. ).

[0120] The quantizer used to perform quantization can be predefined, such as protocol predefined, or it can be indicated by the network side; this application does not limit this.

[0121] In another implementation, the encoder on the terminal side can also compress and quantize the target CSI, outputting CSI feedback information. The decoder on the network side can also dequantize and decompress the CSI feedback information to obtain the reconstructed CSI.

[0122] It should be noted that the encoder on the terminal side can be deployed inside the terminal device or in other devices outside the terminal device, such as the aforementioned OTT host or cloud server; the decoder on the network side can be deployed inside the network device or in other devices outside the network device, such as the aforementioned smart network element.

[0123] It should be understood that although the figure shows an encoder and a decoder, this is only a model division from a functional perspective. The encoder can also be called the first model, and the decoder can also be called the second model. Furthermore, this application does not limit the number of models included in the AE model.

[0124] It should also be understood that the AE model is only one possible model for achieving the above functions and should not constitute any limitation on this application. The AE model can also be replaced by other AI models that can achieve the same or similar functions.

[0125] 3. Model Development: Model development refers to building and training models using AI technology to solve specific inference tasks. For example, model development may include one or more of the following: model training, model adaptation, or model enhancement. Model training includes one or more of the following: initial model training, model retraining, model fine-tuning, and model updating. Initial model training refers to the process of training the AI ​​model for the first time. Model retraining refers to the process of training the model again or multiple times after initial training. Model fine-tuning refers to making minor adjustments to the model parameters based on an existing pre-trained model to adapt to a new task or dataset. Model updating typically refers to improving or upgrading a trained model to enhance its performance or adapt to new data distributions. Model adaptation refers to adapting the model format, that is, converting the received model parameters and / or model files into a model format executable by the local device for inference on that device. Model enhancement refers to improving model functionality through optimization of model structure and training. This includes adding additional features or neural networks to existing model functionalities, enabling the enhanced model to be applicable to more scenarios or richer requirements. Model deployment refers to applying the developed model to real-world scenarios.

[0126] Model development can also be represented as offline engineering in the standard.

[0127] It should be understood that the terms such as model development, model training, model adaptation, and model enhancement used in this document are introduced only to facilitate understanding of offline projects. In specific implementations, the operations involved by these terms may not be clearly distinguished. The operations involved by each term may include, but are not limited to, the contents listed in this document. Alternatively, offline projects may also include some of the operations listed above, or they may include other operations. This application does not limit these operations.

[0128] 4. Model Training: By selecting an appropriate loss function, the model parameters are trained using optimization algorithms to minimize the difference between the model's predicted values ​​and the ground truth (or target values, labels).

[0129] The model training methods involved in this application include supervised learning, self-supervised learning, and knowledge distillation. These methods will be explained in detail below.

[0130] Supervised learning, also known as supervised instruction, involves using machine learning algorithms to learn the mapping relationship between sample values ​​and labels based on collected sample values ​​and labels. This learned mapping relationship is then expressed using a machine learning model. The process of training the machine learning model is the process of learning this mapping relationship. For example, in signal detection, the noisy received signal is the sample, and the corresponding real constellation point is the label. Machine learning aims to learn the mapping relationship between samples and labels through training, that is, to enable the machine learning model to learn a signal detector. During training, the model parameters are optimized by calculating the error between the model's predicted values ​​and the real labels. Once the mapping relationship is learned, it can be used to predict the sample label of each new sample. The mapping relationship learned in supervised learning can include linear and nonlinear mappings. Based on the type of label, the learning task can be divided into classification tasks and regression tasks.

[0131] Self-supervised learning is a type of unsupervised learning. Unsupervised learning relies on collected sample values ​​to allow algorithms to discover inherent patterns within the samples themselves. Self-supervised learning uses the samples themselves as supervisory signals; that is, the model learns the mapping relationship from sample to sample. During training, model parameters are optimized by calculating the error between the model's predicted values ​​and the actual samples. Self-supervised learning can be used in signal compression and decompression recovery applications; common algorithms include autoencoders and generative adversarial networks.

[0132] Knowledge distillation: Generally, large models are often single complex networks or collections of networks, possessing excellent performance and generalization ability, while small models, due to their smaller network size, have limited expressive power. Therefore, the knowledge learned by the large model can be used to guide the training of the small model; this process is called knowledge distillation. Knowledge distillation can enable small models to achieve performance comparable to large models, but with fewer parameters and shorter inference latency, thus achieving model compression and acceleration. Furthermore, directly training small models with massive amounts of data often does not yield good performance, while training large models with massive amounts of data and then using the large model to perform knowledge distillation on the small model can achieve better continuation results. In addition, knowledge distillation can also be used to integrate and transfer datasets from different domains.

[0133] Knowledge distillation employs a teacher-student model, where a teacher model assists in training a student model. The teacher model is a complex, large model, while the student model is a simple, small model. Because the teacher model has strong learning capabilities, it can transfer the knowledge it learns to the relatively weaker student model, thereby enhancing the student model's generalization ability. The complex, cumbersome, but effective teacher model remains offline, simply acting as a mentor; the flexible and lightweight student model is the one actually deployed for prediction tasks.

[0134] 5. Performance metric function: This function describes the difference between the model's predicted values ​​and the actual values. Model training can end when the value of the performance metric function reaches the range required by the performance indicator.

[0135] The performance metric function can be a loss function, also known as an objective function or cost function, used to measure the difference between the predicted and actual values. A higher output value (loss) indicates a greater difference, and model training can be a process of minimizing this difference. Loss functions involved in the embodiments of this application include, for example, mean absolute error (MAE), mean squared error (MSE), and normalized mean squared error (NMSE). MAE can also be called L1 loss.

[0136] Performance metrics can also be used to measure the similarity between predicted and actual values. Examples of performance metrics used to measure the similarity between predicted and actual values ​​include generalized cosine similarity (GCS) and square generalized cosine similarity (SGCS).

[0137] The calculation methods for each performance metric function can be found in the formulas below:

[0138] in, These are predicted values, including N values: H represents the true values, including N values: w1, w2, ..., w N .

[0139] In another implementation, the above GCS and SGCS can also be replaced by the following loss functions: the difference between GCS and 1, and the difference between SGCS and 1. In this case, the model training process is the process of reducing this difference.

[0140] 6. Model Files and Model Parameters: Model files and / or model parameters can be used to define the model. Model files can indicate the model structure, which may include, but is not limited to, feedforward neural networks (FNNs), convolutional neural networks (CNNs), and recurrent neural networks (RNNs). The model file can have a fixed format, such as a standard predefined format or a format pre-negotiated by both ends of the connection. Model parameters can refer to parameters in the neural network model, such as, but not limited to, the number of layers in the neural network, the type and weights of neurons in each layer, etc. This application does not limit the method of distributing the parameters of the reference model.

[0141] Take DNN as an example. The idea behind DNN comes from the neuronal structure of the brain. Each neuron can perform a weighted summation operation on its inputs and then use the result of the weighted summation operation to generate an output through a nonlinear function. Figure 6 shows an example of a neuron structure. The input of the neuron shown in Figure 6 is x = [x0 x1 … x N-1 The weights corresponding to the inputs are w = [w0 w1 … w] N-1 The bias of the weighted summation is b. The nonlinear function f() can take many forms; for example, the nonlinear function f() can be the maximum value function max{0, x}. Then the effect of a neuron's execution is... Where N is a positive integer; n is a positive integer greater than or equal to 0 and less than or equal to (N-1).

[0142] A DNN typically has multiple neural network layers, including an input layer, one or more hidden layers, and an output layer. Generally, the first layer is the input layer, the last layer is the output layer, and the layers in between are hidden layers. Each layer contains multiple neurons. Layers are fully connected; that is, any neuron in the i-th layer is connected to any neuron in the (i+1)-th layer. The input layer processes the received numerical values ​​(i.e., the DNN's input) through neurons and then passes them to the hidden layers. Similarly, the hidden layers pass the computation results to the final output layer, producing the DNN's output. Figure 7 shows an example of a DNN. The DNN model shown in Figure 7 has three neural network layers: an input layer, a hidden layer, and an output layer.

[0143] It should be understood that the examples in conjunction with Figures 6 and 7 above are shown for ease of understanding only and should not constitute any limitation on this application. This application does not limit the structure and parameters used in the AI ​​model.

[0144] One of the model structure or model parameters can be predefined, while the other can be sent by the sender (e.g., the network side). Alternatively, both the model structure and model parameters can be sent by the sender (e.g., the network side). This application does not impose any restrictions on this.

[0145] In this embodiment of the application, sending a model may refer to sending a model file and / or model parameters, and receiving a model may refer to receiving a model file and / or model parameters.

[0146] 7. Two-way connection: This refers to the connection between the sender (e.g., the network side) and the receiver (e.g., the terminal side). It can be a connection between datasets or between models. A dataset is used in machine learning for model training, validation, and testing; the quantity and quality of the dataset affect the effectiveness of machine learning. In this embodiment, the dataset can be used for model training.

[0147] The following section uses CSI compressed inference tasks as an example to explain the connection between the dataset and the model.

[0148] Dataset integration mainly refers to the sender providing a dataset to the receiver for model training. Figure 8 is a schematic diagram of dataset integration between the network side and the terminal side.

[0149] As shown in Figure 8, the network can first acquire the dataset through joint training. For example, the network can use a virtual CSI compression model (e.g., an encoder) to compress the target CSI, and then use a CSI reconstruction model (e.g., a decoder) to decompress the target CSI. That is, the target CSI is the input to the CSI compression model, and the output of the CSI compression model can be the compressed CSI. This compressed CSI can be the input to the CSI reconstruction model, and the CSI reconstruction model can output the reconstructed CSI. In this way, the dataset can include one or more of the following: target CSI or reconstructed CSI.

[0150] Optionally, the compressed CSI can be further quantized to obtain CSI feedback information, which can be used for CSI report submission in specific implementations. Accordingly, the dataset may also include one or more of the following: target CSI, CSI feedback information, or reconstructed CSI. The quantizer can be predefined, such as protocol-predefined, or it can be issued by the sender. Optionally, the dataset also includes: a quantizer. The quantizer can also be replaced by a quantization method.

[0151] The network side can distribute this dataset to the terminal side. The terminal side can then train a model based on the received dataset to obtain a CSI compressed model for the terminal side.

[0152] Model interoperability mainly refers to the sender providing model files and / or model parameters to the receiver for model deployment, development (e.g., model training), and deployment. In this embodiment, the sender can provide model files and / or model parameters to the receiver for model training. Figure 9 is a schematic diagram of model interoperability between the network side and the terminal side.

[0153] As shown in Figure 9, the network side can first obtain the CSI compressed model through joint training. The specific process of network-side joint training can be found in the explanation above combined with Figure 8, and will not be repeated here. The network side can then distribute the CSI compressed model or CSI reconstructed model obtained through joint training to the terminal side. For example, the network side can send the model file and / or model parameters of the CSI compressed model to the terminal side, or send the model file and / or model parameters of the CSI reconstructed model to the terminal side. The terminal side can then train the model based on the received model file and / or model parameters.

[0154] It should be understood that the CSI compression model shown above in conjunction with Figures 8 and 9 can be an encoder, and the CSI reconstruction model can be a decoder.

[0155] 8. Dataset: Data used for model training, validation, or testing in machine learning. The quantity and quality of the data affect the effectiveness of machine learning. Training data can include the input of the AI ​​model, or the input and target output of the AI ​​model. The target output is the target value of the AI ​​model's output, also known as the ground truth, comparison ground truth, label, or labeled sample.

[0156] In practical implementation, the device acquiring and / or managing the dataset may be different from the device training the AI ​​model. Therefore, the device acquiring and / or managing the dataset needs to send the dataset to the device training the AI ​​model so that the latter can develop the model based on the received dataset. For example, the network side can provide the dataset to the terminal side for model development and deployment. Therefore, how to transmit datasets while reducing air interface overhead has become a pressing technical problem.

[0157] In view of this, this application provides a method that associates updated data with previously transmitted data, thereby eliminating the need to transmit a dataset containing both the updated and previously transmitted data. This reduces air interface overhead and avoids redundant training for the dataset receiver.

[0158] The communication method provided in this application embodiment will be described below with reference to the accompanying drawings, taking the interaction between various devices as an example.

[0159] It should be noted that the first or second device in the following embodiments can be replaced by a device on the terminal device side, a device on the network device side, or a device on the core network (CN) side.

[0160] The devices on the terminal device side can include the terminal device itself, the communication module within the terminal device, or the circuits or chips within the terminal device responsible for communication functions (such as a modem chip, also known as a baseband chip, or a SoC chip or SIP chip containing a modem core, etc.), or can be included in the AI ​​entities on the terminal device side. The AI ​​entities on the terminal device side can be the terminal device itself, or they can be AI entities that serve the terminal device, such as servers, such as OTT servers or cloud servers.

[0161] Network device-side equipment can include the network device itself, communication modules within the network device, or circuits or chips within the network device responsible for communication functions (such as modem chips, also known as baseband chips, or system-on-a-chip (SoC) chips or SIP chips containing modem cores, etc.), or it can include AI entities on the network device side. AI entities on the network device side can be the network device itself, or AI entities serving the network device, such as RICs, OAMs, or servers, such as OTT servers or cloud servers.

[0162] The equipment on the core network side may include the core network element itself, the communication module within the core network element, or the circuit or chip responsible for communication functions within the core network element (such as a modem chip, also known as a baseband chip, or a system-on-a-chip (SoC) chip or SIP chip containing a modem core, etc.), or may include AI entities on the core network side. The AI ​​entities on the core network side can be the core network element itself, or they can be AI entities serving the core network element, such as servers, such as OTT servers or cloud servers. The method provided in this application will be described in detail below with reference to the accompanying drawings.

[0163] Figure 10 is a schematic flowchart of a communication method 1000 provided in an embodiment of this application. The steps of method 1000 will be described in detail below.

[0164] S1010, the first device sends the first instruction information.

[0165] Correspondingly, the second device receives the first instruction information.

[0166] The first instruction information is used to indicate the identifier of the dataset that the second device reports.

[0167] In one possible implementation, in S1010, the first indication information sent by the first device to the second device can be replaced by request message #1, which is used to request the identifier of the dataset supported by the second device.

[0168] For example, in the scenario of docking two AI models, request message #1 can be called docking request.

[0169] For example, if the second dataset that the first device is about to send to the second device is associated with the first dataset, then request message #1 can be used to request the second device to confirm whether the first dataset is supported.

[0170] It should be noted that S1010 is an optional step. For example, if the second device periodically reports the identifiers of the datasets supported by the second device to the first device, then method 1000 may not include S1010. Or, for example, if all the datasets supported by the second device originate from the first device, then method 1000 may not include S1010.

[0171] S1020, the second device sends the first request message.

[0172] Accordingly, the first device receives the first request message.

[0173] The first request message is used to request at least one dataset #1, which is used for model development.

[0174] For example, at least one dataset #1 requested by the first request message is used for model development of the first model or initial model described in S1060 below.

[0175] For example, in a scenario where two AI models are connected, the first request message can be called a connection request.

[0176] S1030, the second device sends the identifier of at least one dataset supported by the second device.

[0177] Accordingly, the first device receives the identifier of at least one dataset supported by the second device.

[0178] The identifier of at least one dataset includes the identifier of the first dataset.

[0179] Taking the first dataset from at least one dataset as an example, if the second device supports the first dataset, the second device can use the first dataset for model development. Model development may include one or more of the following: model training, offline engineering, or model augmentation. Model training may include initial model training and / or model retraining. For a more detailed description of model development, please refer to the terminology section above.

[0180] For example, if method 1000 executes S1010, the second device may, in response to the first instruction information, send the identifier of at least one dataset supported by the second device to the first device.

[0181] For example, if method 1000 does not execute S1010, the second device may periodically send the identifier of at least one dataset supported by the second device to the first device. Alternatively, the second device may proactively send the identifier of at least one dataset supported by the second device to the first device when there is a need for model development.

[0182] In one possible implementation, if the second device receives a request message #1, which requests the second device to determine whether it supports the first dataset, then in S1030, the second device sends a first response message to the first device. The first response message indicates that the second device supports the first dataset, or indicates that the second device does not support the first dataset.

[0183] It should be noted that S1030 is an optional step. For example, if the datasets supported by the second device all come from the first device, then method 1000 may not include S1030.

[0184] It should also be noted that if method 1000 includes S1020 and S1030, then S1020 and S1030 can be the same step. For example, the first request message sent by the second device to the first device includes an identifier of at least one dataset supported by the second device.

[0185] S1040, the first device sends the first information.

[0186] Correspondingly, the second device receives the first information.

[0187] The first piece of information indicates the association between at least two identifiers. This association between at least two identifiers includes the association between the first identifier and identifiers in the first dataset.

[0188] It should be noted that S1040 is an optional step. For example, if the first information is pre-configured or predefined by the protocol, then method 1000 may not include S1040. Alternatively, if, in S1050 below, the first identifier sent by the first device is the same as the identifier of the first dataset, then method 1000 may not include S1040.

[0189] S1050, the first device sends the second dataset and the first identifier.

[0190] Correspondingly, the second device receives the second dataset and the first identifier.

[0191] The first identifier can also be called the identifier of the second dataset.

[0192] In one possible implementation, the first identifier is the same as the identifier of the first dataset.

[0193] In one possible implementation, the first identifier is associated with the identifier of the first dataset.

[0194] It should be understood that if the first device determines that the second dataset is associated with the first dataset in at least one dataset supported by the second device, then the first identifier sent by the first device is the same as the identifier of the first dataset, or is associated with the identifier of the first dataset.

[0195] S1060, the second device performs model development on the first model or the initial model.

[0196] Specifically, in S1060, the second device performs model development on the first model or initial model based on at least one dataset to obtain a second model. The at least one dataset includes either the first dataset or the second dataset, or the at least one dataset includes the second dataset.

[0197] It should be understood that after the second device receives the second dataset and the first identifier, if the first identifier is the same as the identifier of the first dataset, or if the first identifier is associated with the identifier of the first dataset, then the second device will develop a second model based on at least one dataset using the first model or the initial model.

[0198] The first model is determined based on the second dataset and the initial model. In other words, the first model is obtained by the second device from the initial model based on the second dataset.

[0199] The initial model may include model parameters and / or a model file. The model file can be used to indicate the model structure, which may include, but is not limited to, FNN, CNN, and RNN. The model file may have a fixed format, such as a standard predefined format, or a format pre-negotiated by the two endpoints. Model parameters may refer to parameters in the neural network model, such as, but not limited to, the number of layers in the neural network, the type and weights of neurons in each layer, etc. The initial model may be sent from the first device to the second device, or it may be pre-configured or predefined by a protocol; this application does not limit this.

[0200] This application does not limit the method by which the second device develops a model based on at least one dataset.

[0201] In one possible implementation, the second device performs initial model training or offline engineering on the initial model based on the first and second datasets to obtain the second model.

[0202] In one possible implementation, the second device performs initial model training or offline engineering on the initial model based on the second dataset to obtain the second model.

[0203] In one possible implementation, the second device performs offline engineering, model retraining, model fine-tuning, model updating, or model enhancement on the first model based on the first and second datasets to obtain the second model.

[0204] In one possible implementation, the second device performs offline engineering, model retraining, model fine-tuning, model updating, or model enhancement on the first model based on the second dataset to obtain the second model.

[0205] In this embodiment, if the first device needs to send a supplementary second dataset to the second device based on the first dataset, so that the second device can use the second dataset to develop a first model or initial model associated with the first dataset, the first device can use the identifier of the first dataset as the identifier of the second dataset, or associate the identifier of the first dataset with the identifier of the second dataset. This helps reduce overhead and avoids redundant training by the second device. For example, since the identifier of the first dataset is the same as or associated with the identifier of the second dataset, the first device does not need to send both the first and second datasets to the second device simultaneously. Instead, by sending the second dataset to the second device, the second device can use both the second dataset and the previously received first dataset for model development. Similarly, since the identifier of the first dataset is the same as or associated with the identifier of the second dataset, by sending the second dataset to the second device, the second device can use the second dataset to develop the first model, thus avoiding redundant training using the first dataset.

[0206] It should be understood that in S1060, during the process of the second device developing the first model or the initial model, the second device can monitor the model development process based on the performance indicators corresponding to the dataset, so that the second model can meet the performance indicators corresponding to the dataset.

[0207] The following section, with reference to Figure 11, explains how the second device determines the performance indicators that the second model should meet.

[0208] Figure 11 is a schematic flowchart of a communication method 1100 provided in an embodiment of this application. The steps of method 1100 will be described in detail below.

[0209] S1101, the first device sends the first dataset and the identifier of the first dataset.

[0210] Correspondingly, the second device receives the first dataset and its identifier.

[0211] S1102, the first device sends the first performance indicator.

[0212] Correspondingly, the second device receives the first performance indicator.

[0213] The first performance metric corresponds to the first dataset.

[0214] For example, the performance metric can specifically be the range of performance requirements that the values ​​of one or more loss functions should meet, such as the accuracy range. This loss function may include, but is not limited to, GCS, SGCS, MSE, NMSE, MAE, etc.

[0215] It should be understood that after receiving the first dataset and the first performance metric, the second device can develop a first model based on the initial model using the first dataset and the first performance metric, and the first model satisfies the first performance metric. For example, if the first model satisfies the first performance metric, the error of the inference result obtained when performing an inference task based on the first model is within the error range corresponding to the first performance metric. For instance, if the first performance metric is MSE#1, then if the first model satisfies the first performance metric, the MSE of the inference result obtained when performing an inference task based on the first model is less than or equal to MSE#1. As another example, if the first performance metric includes SGCS#1 and SGCS#2, and SGCS#1 is greater than SGCS#2, then if the first model satisfies the first performance metric, the SGCS of the inference result obtained when performing an inference task based on the first model is between SGCS#1 and SGCS#2.

[0216] It should be noted that S1101 and S1102 can be the same step. In other words, the first dataset, the identifier of the first dataset, and the first performance metric sent by the first device to the second device are carried in the same message.

[0217] It should also be noted that S1102 can be an optional step. For example, if the first performance metric corresponding to the first dataset is predefined by the protocol, then method 1100 may not include S1102.

[0218] It should also be noted that Figure 11 illustrates the example of the first device sending the first dataset, the identifier of the first dataset, and the first performance index to the second device. In actual implementation, it may be other devices, different from the first device, that send the first dataset, the identifier of the first dataset, and the first performance index to the second device.

[0219] S1103, the first device sends the first instruction information.

[0220] Correspondingly, the second device receives the first instruction information.

[0221] S1104, the second device sends the first request message.

[0222] Accordingly, the first device receives the first request message.

[0223] S1105, the second device sends the identifier of at least one dataset supported by the second device.

[0224] Accordingly, the first device receives the identifier of at least one dataset supported by the second device.

[0225] S1106, the first device sends the first information.

[0226] Correspondingly, the second device receives the first information.

[0227] S1107, the first device sends the second dataset and the first identifier.

[0228] Correspondingly, the second device receives the second dataset and the first identifier.

[0229] S1103 to S1107 can be referenced from S1010 to S1050 in Method 1000 above.

[0230] S1108, the first device sends the second performance indicator.

[0231] Correspondingly, the second device receives the second performance index.

[0232] The second performance metric corresponds to the first dataset. The second performance metric may differ from the first performance metric, or the second performance metric may differ from the first performance metric; this application does not impose any limitations on this.

[0233] Optionally, in S1108, the first device may also send an identifier of the first dataset so that the second device can determine that the second performance metric corresponds to the first dataset.

[0234] It should be understood that if the protocol predefines or the second device preconfigures the correspondence between the identifiers of different datasets and different performance indicators, the first device may not send the identifier of the first dataset.

[0235] It should also be understood that if the first identifier is the same as the identifier of the first dataset, then if the second performance metric corresponds to the first dataset, the second performance metric also corresponds to the second dataset.

[0236] It should be noted that S1108 is an optional step. For example, if the first device does not need to update the performance metrics corresponding to the first dataset, then method 1100 may not include S1108.

[0237] S1109, the first device sends the third performance indicator.

[0238] Correspondingly, the second device receives the third performance indicator.

[0239] The third performance metric corresponds to the second dataset.

[0240] In one possible implementation, S1109 and S1107 are the same step; in other words, the second dataset, the first identifier, and the third performance metric sent by the first device to the second device are carried in the same message. Correspondingly, the second dataset, the first identifier, and the third performance metric received by the second device are also carried in the same message, thus allowing the second device to determine that the third performance metric corresponds to the second dataset.

[0241] In one possible implementation, S1109 and S1107 are different steps. In S1109, the first device can also send a second identifier to the second device so that the second device can determine the correspondence between the third performance index and the second dataset based on the second identifier.

[0242] It should be understood that if the protocol predefines or the second device preconfigures the correspondence between the identifiers of different datasets and different performance indicators, then in S1108, the first device may not send the first identifier.

[0243] It should also be understood that if the first identifier is the same as the identifier of the first dataset, then if the third performance metric corresponds to the second dataset, the third performance metric also corresponds to the first dataset.

[0244] Optionally, if the first identifier is associated with the identifier of the first dataset, then method 1100 may include S1109.

[0245] It should be noted that S1109 is an optional step. For example, if the performance metric for the second dataset is the same as that for the first dataset, then method 1100 may not include S1109.

[0246] S1110, the second device performs model development on the first model or the initial model.

[0247] Specifically, in S1110, the second device performs model development on the first model or the initial model based on at least one dataset and at least one performance metric to obtain a second model.

[0248] For example, at least one dataset includes a first dataset and a second dataset, and at least one performance metric includes a target performance metric corresponding to the first dataset and a target performance metric corresponding to the second dataset.

[0249] For example, at least one dataset includes a second dataset, and at least one performance metric includes a target performance metric corresponding to the second dataset.

[0250] Here, the target performance metric corresponding to the first dataset refers to the performance metric used to monitor the model development process based on the first dataset. In other words, the model determined based on the first dataset will meet the target performance metric corresponding to the first dataset. It can be understood that the first dataset may correspond to multiple performance metrics, but not all performance metrics corresponding to the first dataset are used to monitor the model development process based on the first dataset. In other words, at least one performance metric corresponding to the first dataset includes the target performance metric corresponding to the first dataset.

[0251] Similarly, the target performance metric for the second dataset refers to the performance metric used to monitor the model development process based on the second dataset. In other words, the model determined based on the second dataset will meet the target performance metric corresponding to the second dataset. It can be understood that the first dataset may correspond to multiple performance metrics, but not all performance metrics corresponding to the first dataset are used to monitor the model development process based on the first dataset. In other words, at least one performance metric corresponding to the second dataset includes the target performance metric corresponding to the second dataset.

[0252] It should be understood that after the second device receives the second dataset and the first identifier, if the first identifier is the same as the identifier of the first dataset, or if the first identifier is associated with the identifier of the first dataset, then the second device will develop the first model or the initial model based on at least one dataset and at least one performance metric to obtain the second model.

[0253] The first model is determined based on the second dataset and the initial model. In other words, the first model is obtained by the second device from the initial model based on the second dataset. The initial model can be referred to in the description in S1060 above.

[0254] The following describes how the second device determines the target performance metrics corresponding to the first dataset and / or the second dataset.

[0255] For example, the second device determines the target performance metrics corresponding to the first dataset and / or the second dataset based on one or more of the following rules.

[0256] Rule 1: If multiple performance metrics corresponding to dataset #1 (or the identifier of dataset #1) are obtained, the target performance metric corresponding to dataset #1 is the latest obtained performance metric corresponding to dataset #1 (or the identifier of dataset #1).

[0257] Rule 2: If the identifier of dataset #1 is the same as the identifier of dataset #2, or if they are related, then the target performance metric corresponding to dataset #1 is the same as the target performance metric corresponding to dataset #2.

[0258] Optionally, rule 2 can be replaced with: if the identifier of dataset #1 is the same as the identifier of dataset #2, then the target performance metric corresponding to dataset #1 is the same as the target performance metric corresponding to dataset #2.

[0259] Rule 3: If the identifier of dataset #1 is the same as the identifier of dataset #2, or if they are related, then the target performance metrics corresponding to dataset #1 and dataset #2 are the latest obtained performance metrics corresponding to dataset #1 (or the identifier of dataset #1) and / or dataset #2 (or the identifier of dataset #2).

[0260] Optionally, rule 3 can be replaced with: if the identifier of dataset #1 is the same as the identifier of dataset #2, then the target performance metrics corresponding to dataset #1 and dataset #2 are the latest obtained performance metrics corresponding to dataset #1 (or the identifier of dataset #1) and / or dataset #2 (or the identifier of dataset #2).

[0261] Rule 4: If the identifier of dataset #1 is the same as the identifier of dataset #2, or if they are related, then the target performance metrics corresponding to dataset #1 and dataset #2 shall be determined respectively based on Rule 1.

[0262] Optionally, rule 4 can be replaced with: if the identifier of dataset #1 is related to the identifier of dataset #2, then the target performance index corresponding to dataset #1 and the target performance index corresponding to dataset #2 are determined based on rule 1.

[0263] The following describes how the second device determines the target performance metrics corresponding to the first dataset and / or the second dataset based on the above rules.

[0264] In implementation method 1, if method 1100 executes S1102 but not S1108 and S1109, meaning the second device only obtains the first performance metric corresponding to the first dataset, then based on rule 1 above, the second device determines the target performance metric corresponding to the first dataset as the first performance metric. Furthermore, based on rules 2 and 3 above, the second device determines the target performance metric corresponding to the second dataset as the second performance metric.

[0265] In implementation method 2, if method 1100 executes S1102 and S1108, and S1108 follows S1102 (i.e., the second device obtains the first performance metric and the second performance metric corresponding to the first dataset), then based on rule 1 above, the second device determines the target performance metric corresponding to the first dataset as the newly obtained second performance metric. Furthermore, based on rules 2 and 3 above, the second device determines the target performance metric corresponding to the second dataset as the second performance metric.

[0266] In implementation method 3, if method 1100 executes S1102 and S1109, and S1109 is after S1102, that is, the second device obtains the first performance index corresponding to the first dataset and the third performance index corresponding to the second dataset, then based on the above rules 1 and 4, the second device determines the target performance index corresponding to the first dataset as the first performance index and determines the target performance index corresponding to the second dataset as the second performance index.

[0267] In implementation method 4, if method 1100 executes S1102 and S1109, and S1109 is after S1102, that is, the second device obtains the first performance index corresponding to the first dataset and the third performance index corresponding to the second dataset, then based on the above rule 3, the second device determines that the target performance index corresponding to the first dataset is the same as the target performance index corresponding to the second dataset, and the target performance index corresponding to the first dataset and the second dataset is the third performance index corresponding to the second dataset that the second device has just obtained.

[0268] Optionally, if the first identifier is associated with the identifier of the first dataset, then when method 1100 executes S1102 and S1109, the second device determines the target performance index corresponding to the first dataset and the second dataset according to the above implementation method 3.

[0269] Optionally, if the first identifier is the same as the identifier of the first dataset, then if S1102 and S1109 are executed in method 1100, the second device determines the target performance index corresponding to the first dataset and the second dataset according to the above implementation method 4.

[0270] In implementation method 5, if method 1100 executes S1102, S1108, and S1109, and S1108 and S1109 are after S1102, that is, the second device obtains the first performance index and the second performance index corresponding to the first dataset, and obtains the third performance index corresponding to the second dataset, then based on the above rules 1 and 4, the second device determines the target performance index corresponding to the first dataset as the second performance index, and determines the performance index corresponding to the second dataset as the third performance index.

[0271] In implementation method 6, if method 1100 executes S1102, S1108, and S1109, and S1109 follows S1108 and S1102 (i.e., the second device obtains the first and second performance metrics corresponding to the first dataset, and the third performance metric corresponding to the second dataset), then based on rule 3 above, the second device determines that the target performance metric corresponding to the first dataset is the same as the target performance metric corresponding to the second dataset. If S1109 follows S1108, the second device determines that the target performance metric corresponding to the first and second datasets is the third performance metric most recently obtained by the second device corresponding to the second dataset; or, if S1108 follows S1109, the second device determines that the target performance metric corresponding to the first and second datasets is the second performance metric most recently obtained by the second device corresponding to the first dataset.

[0272] Optionally, if the first identifier is associated with the identifier of the first dataset, then when method 1100 executes S1102, S1108 and S1109, the second device determines the target performance index corresponding to the first dataset and the second dataset according to the above implementation method 5.

[0273] Optionally, if the first identifier is the same as the identifier of the first dataset, then if S1102, S1108 and S1109 are executed in method 1100, the second device determines the target performance index corresponding to the first dataset and the second dataset according to the above implementation method 6.

[0274] In one possible implementation, if the second device develops a model based on the second dataset for the first model or the initial model, then the second device determines the target performance metric corresponding to the second dataset, but not the target performance metric corresponding to the first dataset. If the second device develops a model based on both the first and second datasets, then the second device determines both the target performance metric corresponding to the second dataset and the target performance metric corresponding to the second dataset.

[0275] The embodiments of this application do not limit the method by which the second device develops a model based on at least one dataset and at least one performance metric.

[0276] In one possible implementation, the second device performs initial model training or offline engineering on the initial model based on the first and second datasets to obtain a second model. The second model satisfies the target performance metrics corresponding to the first and second datasets.

[0277] For example, if the target performance metric for both the first and second datasets is the first performance metric, then the second model satisfies the first performance metric. Alternatively, if both the target performance metric for the first and second datasets is the second performance metric, then the second model satisfies the second performance metric. Or, if both the target performance metric for the first and second datasets is the third performance metric, then the second model satisfies the third performance metric. Finally, if the target performance metric for the first dataset is the second performance metric and the target performance metric for the second dataset is the third performance metric, then the second model satisfies both the second and third performance metrics.

[0278] In one possible implementation, the second device performs initial model training or offline engineering on the initial model based on the second dataset to obtain a second model. The second model satisfies the target performance metrics corresponding to the second dataset.

[0279] In one possible implementation, the second device performs offline engineering, model retraining, model fine-tuning, model updating, or model augmentation on the first model based on the first and second datasets to obtain a second model. The second model satisfies the target performance metrics corresponding to the first and second datasets.

[0280] In one possible implementation, the second device performs offline engineering, model retraining, model fine-tuning, model updating, or model enhancement on the first model based on the second dataset to obtain a second model. The second model satisfies the target performance metric corresponding to the second dataset.

[0281] The meaning of the second model satisfying the performance index can be found in the description of the meaning of the first model satisfying the first performance index in S1102 above.

[0282] In this embodiment of the application, if the first device needs to send a supplementary second dataset to the second device based on the first dataset, so that the second device can use the second dataset to develop the first model or initial model associated with the first dataset, then by defining the method by which the second device determines the performance metrics used to monitor the model development process based on the second dataset, the second device can avoid using incorrect performance metrics to monitor the model development process based on the second dataset, thereby affecting the performance of the model.

[0283] As described above, the first device can be replaced by at least one of a device (such as a terminal device, access network device, or core network element) or an AI entity on the device side. If the first device is replaced by a device and an AI entity on the device side, the following steps in the above embodiments can be performed by the AI ​​entity of the device: the step of sending the dataset and the step of sending performance indicators.

[0284] The second device can be replaced by at least one of a device (such as a terminal device, access network device, or core network element) or an AI entity on the device side. If the first device is replaced by a device and an AI entity on the device side, the following steps in the above embodiments can be performed by the AI ​​entity of the device: the step of receiving the dataset and the step of receiving performance indicators.

[0285] For example, Figure 12 illustrates a schematic flowchart of the method provided in this application from the perspective of the interaction between intelligent network elements, network devices, terminal devices, and OTT servers. The intelligent network element can be an AI entity on the network device side. The OTT server can be an AI entity on the terminal device side.

[0286] As shown in Figure 12, method 1200 may include the following steps.

[0287] S1201, The terminal device sends the identifier of at least one dataset supported by the terminal device.

[0288] Accordingly, the network device receives the identifier of at least one dataset supported by the terminal device.

[0289] For a more detailed description of S1201, please refer to S1030 in Method 1000 above.

[0290] Optionally, in S1201, the OTT server may send the identifier of at least one dataset supported by the OTT server to the terminal device, and then the terminal device may send the identifier of at least one dataset supported by the OTT server to the network device.

[0291] Optionally, if the second dataset is sent by an intelligent network element, method 1200 further includes S1202.

[0292] S1202, The network device sends an identifier of at least one dataset supported by the terminal device.

[0293] Correspondingly, the intelligent network element receives the identifier of at least one dataset supported by the terminal device.

[0294] Optionally, if the identifier of the second dataset sent by the intelligent network element (i.e., the first identifier) ​​is associated with the identifier of the second dataset, then method 1200 further includes S1203.

[0295] S1203, the intelligent network element sends the first information.

[0296] Accordingly, the network device receives the first information.

[0297] For a more detailed description of the first information, please refer to S1040 in Method 1000 above.

[0298] S1204, the network device sends the first message.

[0299] Accordingly, the terminal device receives the first information.

[0300] Optionally, if the recipient of the second dataset is an OTT server, then method 1200 further includes S1205.

[0301] S1205, the terminal device sends the first information.

[0302] Accordingly, the OTT server receives the first information.

[0303] It should be understood that if the first information is pre-configured or predefined by the protocol, then method 1200 may not include S1203 to S1205.

[0304] Optionally, if the second dataset is sent by an intelligent network element, then method 1200 continues to execute S1206.

[0305] S1206, the intelligent network element sends the second dataset and the first identifier.

[0306] Accordingly, the network device receives the second dataset and the first identifier.

[0307] S1207, the network device sends the second dataset and the first identifier.

[0308] Accordingly, the terminal device receives the second dataset and the first identifier.

[0309] Optionally, if the recipient of the second dataset is an OTT server, then method 1200 further includes S1208.

[0310] S1208, the terminal device sends the second dataset and the first identifier.

[0311] Accordingly, the OTT server receives the second dataset and the first identifier.

[0312] For further descriptions of S1206 to S1208, please refer to S1050 in Method 1000 above.

[0313] Furthermore, method 1200 executes S1209 or S1210.

[0314] S1209, The terminal device performs model development on the first model or the initial model.

[0315] S1210, the OTT server performs model development on the first model or initial model.

[0316] For further description of S1209 or S1210, please refer to S1060 in Method 1000 above.

[0317] The methods provided in the embodiments of this application have been described in detail above with reference to several accompanying drawings. The apparatus provided in the embodiments of this application will now be described with reference to the accompanying drawings.

[0318] Figures 13 and 14 are schematic block diagrams of possible apparatuses provided in embodiments of this application. These apparatuses can be used to implement the functions of the first or second apparatus in the above method embodiments, and thus can also achieve the beneficial effects of the above method embodiments.

[0319] Figure 13 is a schematic block diagram of an apparatus provided in an embodiment of this application. The apparatus 1300 shown in Figure 13 may include a processing module 1310 and a communication module 1320.

[0320] In one possible design, device 1300 can be used to implement the communication method implemented by the first device in any of the embodiments shown in Figures 10, 11, and 12. For example, processing module 1310 is used to implement processing-related steps performed by the first device in each method embodiment; communication module 1320 is used to implement sending and / or receiving steps performed by the first device in each method embodiment, such as sending a second dataset, sending a second performance indicator, sending a third performance indicator, sending first information, or receiving one or more of the identifiers of at least one dataset.

[0321] For example, the communication module 1320 is used to send a second dataset and a first identifier, the first identifier being the same as or associated with the identifier of the first dataset; the first dataset is used for model development of a first model or an initial model, the model development including one or more of the following: model training, offline engineering or model enhancement, the first model being determined based on the first dataset and the initial model.

[0322] Optionally, the communication module 1320 is also configured to receive an identifier of at least one dataset, the identifier of the at least one dataset including the identifier of a first dataset.

[0323] Optionally, the communication module 1320 is further configured to send first information, the first information indicating the association relationship between at least two identifiers, the association relationship between the first identifier and the identifier of the first dataset.

[0324] Optionally, the first dataset corresponds to a first performance metric, and the communication module 1320 is also used to send a second performance metric, which corresponds to the first dataset and is different from the first performance metric.

[0325] Optionally, the first dataset corresponds to the first performance metric, and the communication module 1320 is also used to send a third performance metric, which corresponds to the second dataset.

[0326] A more detailed description of the processing module 1310 and the communication module 1320 can be obtained directly from the relevant descriptions in the method embodiments shown in Figures 10, 11 and 12, and will not be repeated here.

[0327] In another possible design, device 1300 can be used to implement the communication method implemented by the second device in any of the embodiments shown in Figures 10, 11, and 12. For example, processing module 1310 is used to implement processing-related steps performed by the second device in each method embodiment, such as model development for a first model or an initial model; communication module 1320 is used to implement sending and / or receiving steps performed by the second device in each method embodiment, such as receiving a second dataset, receiving a second performance indicator, receiving a third performance indicator, receiving first information, or sending one or more of the identifiers of at least one dataset.

[0328] For example, the communication module 1320 is used to receive a second dataset and a first identifier; if the first identifier is the same as or associated with the identifier of the first dataset, then the processing module 1310 is used to perform model development on the first model or the initial model based on at least one dataset to obtain a second model; wherein, at least one dataset includes the second dataset, or includes the first dataset and the second dataset; the first model is determined based on the first dataset and the initial model, and the model development includes one or more of the following: model training, offline engineering, or model enhancement.

[0329] Optionally, the communication module 1320 is also used to send an identifier of at least one dataset, the identifier of the at least one dataset including the identifier of a first dataset.

[0330] Optionally, the communication module 1320 is further configured to receive first information, the first information indicating the association relationship between at least two identifiers, the association relationship between the first identifier and the identifier of the first dataset.

[0331] Optionally, the first dataset corresponds to the first performance index, and the processing module 1310 is further used to determine the second dataset corresponds to the first performance index; the processing module 1310 is specifically used to develop a second model based on at least one dataset and the first performance index to obtain a second model from the first model or the initial model, and the second model satisfies the first performance index.

[0332] Optionally, the first dataset corresponds to the first performance metric, and the communication module 1320 is further configured to receive a second performance metric, which corresponds to the first dataset and is different from the first performance metric; the processing module 1310 is further configured to determine that the second dataset corresponds to the second performance metric; the processing module 1310 is specifically configured to develop a second model based on at least one dataset and the second performance metric to obtain a second model from the first model or the initial model, and the second model satisfies the second performance metric.

[0333] Optionally, the first dataset corresponds to a first performance metric, and the communication module 1320 is further configured to receive a third performance metric, which corresponds to a second dataset; the processing module 1310 is further configured to determine that the first dataset corresponds to a third performance metric; the processing module 1310 is specifically configured to develop a second model based on at least one dataset and a third performance metric, and the second model satisfies the third performance metric.

[0334] Optionally, the first dataset corresponds to a first performance metric. The communication module 1320 is further configured to receive a second performance metric, which corresponds to the first dataset and is different from the first performance metric. The communication module 1320 is also configured to receive a third performance metric, which corresponds to the second dataset. The processing module 1310 is specifically configured to develop a second model based on at least one dataset and at least one performance metric, using the first model or initial model. The second model satisfies at least one performance metric. Wherein, at least one dataset includes the second dataset, and at least one performance metric includes the second performance metric; or, at least one dataset includes both the first and second datasets, and at least one performance metric includes both the second and third performance metric.

[0335] A more detailed description of the processing module 1310 and the communication module 1320 can be obtained directly from the relevant descriptions in the method embodiments shown in Figures 10, 11 and 12, and will not be repeated here.

[0336] It should be noted that the communication module can also be called a transceiver module, transceiver unit, transceiver, transceiver device, or transceiver apparatus, etc. The processing module can also be called a processor, processing board, processing unit, or processing apparatus, etc. Optionally, the communication module is used to execute the sending and receiving operations of the first or second communication device in the above method. The device in the communication module that implements the receiving function can be considered as the receiving module, and the device in the communication module that implements the sending function can be considered as the sending module; that is, the communication module can include both a receiving module and a sending module.

[0337] It should also be noted that, in one possible design, the aforementioned processing module and / or communication module can be implemented through virtual modules. For example, the processing module can be implemented through software functional units or virtual devices, and the communication module can be implemented through software functions or virtual devices. In another possible design, the processing module or communication module can also be implemented through physical devices. For example, if the device is implemented using a chip / chip circuit, the communication module can be an input / output circuit and / or a communication interface, performing input operations (corresponding to the aforementioned receiving operation) and output operations (corresponding to the aforementioned sending operation); the processing module can be an integrated processor, a microprocessor, or an integrated circuit.

[0338] The module division in this embodiment is illustrative and represents only one logical functional division; in actual implementation, other division methods may be used. Furthermore, the functional modules in the various examples of this embodiment can be integrated into a single processor, exist as separate physical entities, or be integrated into a single module. The integrated modules described above can be implemented in hardware or as software functional modules.

[0339] Figure 14 is a schematic diagram of the structure of a communication device provided in another embodiment of this application. As shown in Figure 14, the device 1400 includes a processing circuit 1410 and a communication circuit 1420. The processing circuit 1410 and the communication circuit 1420 are coupled to each other.

[0340] It can be understood that the processing circuit 1410 can be one or more processors, or it can be all or part of the circuits with processing functions in one or more processors.

[0341] Understandably, the communication circuit 1420 can be a transceiver or an input / output interface.

[0342] Optionally, the device 1400 may further include a memory 1430 for storing instructions executed by the processing circuit 1410, or storing input data required for the running instructions of the processing circuit 1410, or storing data generated after the running instructions of the processing circuit 1410.

[0343] It is understood that the memory 1430 may be located outside the processing circuit 1410, or inside the processing circuit 1410.

[0344] As an example, the processing circuit 1410 is used to implement the functions of the processing module 1310, and the communication circuit 1420 is used to implement the functions of the communication module 1320.

[0345] As an example, device 1400 can be a communication device or a chip used in a communication device.

[0346] When device 1400 is a communication device, the communication circuit can be a transceiver; when device 1400 is a chip, the communication circuit can be an input / output circuit, a bus, pins, or other types of communication interfaces. The input circuit in the input / output circuit can be used for receiving, and the output interface can be used for transmitting.

[0347] In one possible implementation, the apparatus 1400 is used to implement the various processes and steps corresponding to the first apparatus in the above method embodiments. In another possible implementation, the apparatus 1400 is used to implement the various processes and steps corresponding to the second apparatus in the above method embodiments.

[0348] It is understood that device 1400 may specifically be the first device or the second device in the above embodiments, or it may be a chip or a chip system. Correspondingly, the communication circuit may be the interface circuit of the chip, or an input / output circuit, which is not limited here. Specifically, device 1400 may be used to execute the various steps and / or processes corresponding to the first device or the second device in the above method embodiments.

[0349] When the aforementioned communication device is a chip or OTT device applied to the first device, the chip or OTT device of the first device implements the functions of the first device in the above method embodiments, for example, implementing the processing functions of the first device. The chip or OTT device of the first device receiving information from the second device can be understood as the information being first received by other modules (such as radio frequency modules or antennas) in the first device, and then sent by these modules to the chip or OTT device of the first device. The chip or OTT device of the first device sending information to the second device can be understood as the information being first sent by the chip or OTT device of the first device to other modules (such as radio frequency modules or antennas) in the first device, and then sent by these modules to the second device.

[0350] When the aforementioned communication device is a chip or OTT device applied to the second device, the chip or OTT device of the second device implements the functions of the second device in the above method embodiments, for example, implementing the processing functions of the second device. The chip or OTT device of the second device receiving information from the first device can be understood as the information being first received by other modules (such as radio frequency modules or antennas) in the second device, and then sent by these modules to the chip or OTT device of the second device. The chip or OTT device of the second device sending information to the first device can be understood as the information being first sent by the chip or OTT device of the second device to other modules (such as radio frequency modules or antennas) in the second device, and then sent by these modules to the first device.

[0351] This application also provides a computer program product that, when run on a processor, can implement the communication method executed by the first device or the second device in the above method embodiments.

[0352] This application also provides a computer-readable storage medium containing computer instructions that, when executed on a processor, can implement the communication method performed by the first or second device in the above method embodiments.

[0353] This application also provides a communication system, including the aforementioned first device and second device. The first device can be used to implement the communication method implemented by the first device in the above method embodiments, and the second device can be used to implement the communication method implemented by the second device in the above method embodiments.

[0354] It is understood that the processor in the embodiments of this application may be any of the following devices or all or part of the circuitry used for processing functions: a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), processors for AI, field-programmable gate arrays (FPGAs), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. A general-purpose processor may be a microprocessor or any conventional processor.

[0355] The processor used for AI can be one or more of the following: graphics processing unit (GPU), neural processing unit (NPU), tensor processing unit (TPU), and data processing unit (DPU).

[0356] For example, one possible implementation of a processor for AI could be the AI ​​processor 1500 shown in Figure 15.

[0357] As shown in Figure 15, the AI ​​processor 1500 may include one or more of the following: AI core, digital vision pre-processing (DVPP) module, task scheduler (TS), L3 cache, AI CPU, control CPU, L2 cache, universal serial bus (USB) interface, network card, peripheral component interconnect express (PCIe) interface (PCIe is a high-speed serial computer expansion bus standard), double data rate (DDR) / high bandwidth memory (HBM) interface, generation purpose input / output (GPIO) / inter-integrated circuit (I2C) bus, etc.

[0358] The terms “unit”, “module”, etc., used in this specification may be used to refer to computer-related entities, hardware, firmware, combinations of hardware and software, software, or software in execution.

[0359] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0360] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0361] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0362] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0363] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0364] In the above embodiments, the functions of each functional unit can be implemented entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially in the form of a computer program product. This computer program product includes one or more computer instructions (programs). When the computer program instructions (programs) are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium accessible to a computer or a data storage device such as a server or data center that integrates one or more available media. The available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., digital video discs (DVDs)), or semiconductor media (e.g., solid-state disks (SSDs)).

[0365] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0366] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

A communication method, characterized in that, include: Receive the second dataset and the first identifier; If the first identifier is the same as or associated with the identifier of the first dataset, then a second model is obtained by developing the first model or the initial model based on at least one dataset. Wherein, the at least one dataset includes the second dataset, or includes the first dataset and the second dataset; the first model is determined based on the first dataset and the initial model, and the model development includes one or more of the following: model training, offline engineering, or model enhancement. The method according to claim 1, characterized in that, The method further includes: Send the identifier of at least one dataset, wherein the identifier of the at least one dataset includes the identifier of the first dataset. The method according to claim 1 or 2, characterized in that, The first identifier is associated with the identifier of the first dataset, and the method further includes: Receive first information, the first information indicating the association relationship of at least two identifiers, the association relationship of the at least two identifiers including the association relationship between the first identifier and the identifier of the first dataset. The method according to any one of claims 1 to 3 is characterized in that, The first dataset corresponds to the first performance metric, and the method further includes: Determine that the second dataset corresponds to the first performance metric; A second model is obtained by developing the first model or the initial model based on at least one dataset, including: The second model is obtained by developing the first model or the initial model based on the at least one dataset and the first performance metric, wherein the second model satisfies the first performance metric. The method according to any one of claims 1 to 3 is characterized in that, The first dataset corresponds to the first performance metric, and the method further includes: Receive a second performance metric, which corresponds to the first dataset and is different from the first performance metric; Determine that the second dataset corresponds to the second performance metric; A second model is obtained by developing a first model or an initial model based on at least one dataset, including: The second model is obtained by developing the initial model or the first model based on the at least one dataset and the second performance metric, and the second model satisfies the second performance metric. The method according to any one of claims 1 to 3 is characterized in that, The first dataset corresponds to the first performance metric, and the method further includes: Receive a third performance metric, which corresponds to the second dataset; Determine the first performance metric corresponding to the first dataset; A second model is obtained by developing a first model or an initial model based on at least one dataset, including: The second model is obtained by developing the first model or the initial model based on the at least one dataset and the third performance metric, wherein the second model satisfies the third performance metric. The method according to any one of claims 1 to 3 is characterized in that, The first dataset corresponds to the first performance metric, and the method further includes: Receive a second performance metric, which corresponds to the first dataset, and the second performance metric is different from the first performance metric. Receive a third performance metric, which corresponds to the second dataset; A second model is obtained by developing a first model or an initial model based on at least one dataset, including: The second model is obtained by developing the first model or initial model based on the at least one dataset and at least one performance metric, wherein the second model satisfies the at least one performance metric. Wherein, the at least one dataset includes the second dataset, and the at least one performance metric includes the second performance metric; or, The at least one dataset includes the first dataset and the second dataset, and the at least one performance metric includes the second performance metric and the third performance metric. The method according to any one of claims 4 to 6, characterized in that, The first identifier is the same as the identifier of the first dataset. A communication method, characterized in that, include: Send a second dataset and a first identifier, wherein the first identifier is the same as or associated with the identifier of the first dataset; The first dataset is used for model development of a first model or an initial model, and the model development includes one or more of the following: model training, offline engineering, or model enhancement, wherein the first model is determined based on the first dataset and the initial model. The method according to claim 9, characterized in that, The method further includes: Receive the identifier of at least one dataset, wherein the identifier of the at least one dataset includes the identifier of the first dataset. The method according to claim 9 or 10, characterized in that, The first identifier is associated with the identifier of the first dataset, and the method further includes: Send a first message indicating the association between at least two identifiers, the association between the first identifier and the identifiers of the first dataset. The method according to any one of claims 9 to 11, characterized in that, The first dataset corresponds to the first performance metric, and the method further includes: Send a second performance metric, which corresponds to the first dataset, and the second performance metric is different from the first performance metric. The method according to any one of claims 9 to 12, characterized in that, The first dataset corresponds to the first performance metric, and the method further includes: Send a third performance metric, which corresponds to the second dataset. A communication device, characterized in that, It includes functional modules for implementing the method as described in any one of claims 1 to 13. A communication device, characterized in that, It includes one or more processors and communication circuitry, the communication circuitry being used by the communication device to perform at least one of signal input or output; the one or more processors being used to implement the method as described in any one of claims 1 to 13. A computer-readable storage medium, characterized in that, When the computer program is executed by a processor, the method as described in any one of claims 1 to 13 is performed. A computer program product, characterized in that, Includes a computer program, which, when run, causes the method as described in any one of claims 1 to 13 to be performed.

Images