WO2025232813A1 - Communication method and communication apparatus - Google Patents

Abstract

The present application provides a communication method, the method comprising: a terminal device receiving first indicating information sent by a network device, wherein the first indicating information indicates a first resource configuration, the first resource configuration is one of at least one resource configuration supported by a first artificial intelligence (AI) model, and the first AI model is configured for processing channel information obtained on the basis of the first resource configuration measurement. The described technical solution ensures or improves the performance of the first AI model when the network device performs different resource configurations for the terminal device.

Description

A communication method and a communication device

This application claims priority to Chinese Patent Application No. 202410579828.0, filed on May 10, 2024, entitled "A Communication Method and Communication Device", the entire contents of which are incorporated herein by reference.

Technical Field

This application relates to the field of communication technology, and more specifically, to a communication method and a communication device.

Background Technology

In a communication system, network devices determine the downlink data channel resources, modulation and coding scheme (MCS), and precoding and other relevant downlink channel configuration information for scheduling terminal devices based on downlink channel state information (CSI). Terminal devices calculate the downlink CSI by measuring the downlink reference signal and generate a CSI report, which is then fed back to the network devices.

The introduction of artificial intelligence (AI) into wireless communication has led to the emergence of an AI-based CSI feedback mechanism. AI models possess stronger feature extraction capabilities, enabling them to process channel information more effectively, such as predicting and/or compressing it.

Currently, the configuration of reference signals sent by network devices to terminal devices is flexible and variable, which may cause performance fluctuations in AI models during channel information processing. For example, if the AI model on the terminal device side is an AI CSI compression model, it may lead to performance fluctuations in CSI feedback. Similarly, if the AI model on the terminal device side is an AI CSI prediction model, it may lead to performance fluctuations in CSI prediction.

Summary of the Invention

This application provides a communication method that can guarantee or improve the performance of an AI model when a network device performs different resource configurations to a terminal device.

Firstly, a communication method is provided, which can be executed by a first device. The first device can be a device on the first AI model side, or a chip or circuit of the device on the first AI model side. The device on the first AI model side can be replaced by a device on the terminal device side, which can include at least one of a terminal device or an AI entity on the terminal device side. The AI entity on the terminal device side can be the terminal device itself, or an AI entity serving the terminal device, such as a server, like an over-the-top (OTT) server or a cloud server.

The method includes: receiving first indication information, the first indication information indicating a first resource configuration, the first resource configuration being one of at least one resource configuration supported by a first artificial intelligence (AI) model, wherein the first AI model is used to process channel information measured based on the first resource configuration.

The resource configuration supported by the first AI model can be understood as a resource configuration that does not exceed the capabilities of the first AI model; for example, the resource configuration could be a CSI-RS configuration. It should be noted that, in this embodiment, the terminal device can determine the resource configuration supported by the AI model by consulting the AI model's technical documentation or related instructions.

In the technical solution of this application, the terminal device receives a first instruction information sent by the network device to indicate a first resource configuration. The first resource configuration is a resource configuration supported by a first AI model on the terminal device side, which can guarantee or improve the performance of the AI model when the network device performs different resource configurations to the terminal device.

In conjunction with the first aspect, in some implementations of the first aspect, the resource configuration includes: configuration type, offset of adjacent resources, or number of transmissions. The configuration type includes at least one of the following: periodic configuration, semi-static configuration, or aperiodic configuration. Based on the above technical solution, the performance of the AI model can be guaranteed or improved when the network device performs different resource configurations to the terminal device.

In conjunction with the first aspect, in some implementations of the first aspect, the first AI model is used to process channel information measured based on the first resource configuration, and further includes: determining a first channel report, which is determined based on the processing of the channel information measured based on the first resource configuration by the first AI model. Based on the above technical solution, the terminal device processes the channel information measured based on the first resource configuration using the first AI model and generates a first channel report, which can guarantee or improve the performance of the AI model when the network device performs different resource configurations on the terminal device.

It should be understood that the first channel report can be replaced by the first CSI report, or the first CSI feedback information, or the first CSI compressed information, etc. This application does not impose any limitations on this.

In conjunction with the first aspect, in some implementations of the first aspect, the method further includes: sending first information, which is associated with the at least one resource configuration. Based on the above technical solution, the terminal device sends first information to the network device, the first information being associated with at least one resource configuration supported by the first AI model. The network device selects the first resource configuration from the at least one resource configuration and sends it to the terminal device, which can guarantee or improve the performance of the AI model when the network device provides different resource configurations to the terminal device.

In conjunction with the first aspect, in some implementations of the first aspect, the first information includes the at least one resource configuration. Based on the above technical solution, the terminal device reports at least one resource configuration supported by the first AI model to the network device, and the network device selects the first resource configuration from the at least one resource configuration and sends it to the terminal device, which can guarantee or improve the performance of the AI model when the network device provides different resource configurations to the terminal device.

Optionally, the first information may also include at least one resource configuration supported by the second AI model, wherein the second AI model can be understood as one or more AI models different from the first AI model.

In conjunction with the first aspect, in some implementations of the first aspect, the first information includes the identification information of the first AI model. Based on the above technical solution, the terminal device sends the identification information of the first AI model to the network device. The network device determines at least one resource configuration supported by the first AI model based on the identification information of the first AI model, and selects a first resource configuration from the at least one resource configuration and sends it to the terminal device. This can guarantee or improve the performance of the AI model when the network device performs different resource configurations on the terminal device.

In conjunction with the first aspect, in some implementations of the first aspect, the method further includes: sending second indication information, which indicates the correspondence between the identification information of the first AI model and the at least one resource configuration. Based on the above technical solution, the network device determines at least one resource configuration supported by the first AI model according to the identification information of the first AI model and the second indication information, and selects a first resource configuration from the at least one resource configuration and sends it to the terminal device, which can guarantee or improve the performance of the AI model when the network device performs different resource configurations to the terminal device.

In conjunction with the first aspect, in some implementations of the first aspect, the second instruction information further indicates the correspondence between the identification information of the second AI model and at least one resource configuration supported by the second AI model, wherein the second AI model is different from the first AI model. Based on the above technical solution, the network device determines at least one resource configuration supported by the first AI model according to the identification information of the first AI model and the second instruction information, and selects a first resource configuration from the at least one resource configuration and sends it to the terminal device, which can guarantee or improve the performance of the AI model when the network device performs different resource configurations to the terminal device.

It should be understood that the second AI model can be one or more AI models that are different from the first AI model.

In conjunction with the first aspect, in some implementations of the first aspect, sending the second instruction information includes: sending second information, which includes the second instruction information, and the second information includes any one of the following: model-related information, registration request information, and terminal device capability information. Based on the above technical solution, the performance of the AI model can be guaranteed or improved when the network device performs different resource configurations to the terminal device.

In conjunction with the first aspect, in some implementations of the first aspect, the first information includes functional information of the first AI model, which includes any one of the following:

Time-domain channel state information, CSI prediction post-compression function information,

Spatial and frequency domain CSI compression function information,

Spatial, frequency, and time domain CSI compression function information

Time-domain CSI prediction function information

Beam temporal prediction function information.

Based on the above technical solutions, the performance of AI models can be guaranteed or improved when network devices allocate different resources to terminal devices.

In conjunction with the first aspect, some implementations of the first aspect include: sending a second channel report and third indication information, wherein the third indication information indicates that the second channel report is associated with a third AI model, and the second channel report is generated based on channel information obtained from second resource configuration measurements processed by the third AI model. Based on the above technical solution, the performance of the AI model can be guaranteed or improved when the network device performs different resource configurations to the terminal device.

Optionally, the third instruction information may be the identification information of a third AI model.

Optionally, the third AI model can be the first AI model described above, the second resource configuration can be the first resource configuration described above, and correspondingly, the second channel report can be the first channel report described above. In this case, the terminal device sends the second channel report and the third indication information to the network device, so that the network device can use the AI model corresponding to the third AI model to decompress the second channel report to obtain the recovered channel information.

In conjunction with the first aspect, in some implementations of the first aspect, the method further includes: receiving fourth indication information, the fourth indication information indicating the configuration information of the second channel report, the configuration information of the second channel report including the maximum feedback overhead of the second channel report. Based on the above technical solution, the terminal device matches the configuration information of the channel report sent by the network device in real time, selects the most suitable AI model (e.g., a third AI model), which can guarantee or improve the performance of the AI model when the network device performs different resource configurations to the terminal device, and further guarantee the performance of CSI feedback.

In conjunction with the first aspect, in some implementations of the first aspect, the method further includes: determining the information of the second channel report based on the second resource configuration. Based on the above technical solution, the performance of the AI model can be guaranteed or improved when the network device performs different resource configurations to the terminal device, and the performance of CSI feedback can be further guaranteed.

In conjunction with the first aspect, in certain implementations of the first aspect, the information reported by the second channel includes at least one of the following: the number of resources corresponding to one or more measurement resources, the measurement results corresponding to one or more measurement resources, the time-domain information corresponding to one or more measurement resources, the frequency-domain information corresponding to one or more measurement resources, the spatial-domain information corresponding to one or more measurement resources, and the number of bits corresponding to each resource information in one or more measurement resources. Based on the above technical solution, the performance of the AI model can be guaranteed or improved when the network device performs different resource configurations to the terminal device, and the performance of CSI feedback can be further guaranteed.

Secondly, a communication method is provided, which can be executed by a first device. The first device can be a device on the first AI model side, or a chip or circuit of the device on the first AI model side. The device on the first AI model side can be replaced by a device on the terminal device side. The terminal device side can include at least one of a terminal device or an AI entity on the terminal device side. The AI entity on the terminal device side can be the terminal device itself, or an AI entity serving the terminal device, such as a server, such as an over-the-top (OTT) server or a cloud server.

The method includes: sending a second channel report and a third indication information, wherein the third indication information indicates that the second channel report is associated with a third AI model, and the second channel report is generated by processing channel information obtained from second resource configuration measurements based on the third AI model.

In the technical solution of this application, the terminal device matches the configuration information of the channel report sent by the network device in real time, selects the most suitable AI model (e.g., the third AI model), and sends third indication information to the network device so that the network device can use the AI model corresponding to the third AI model to decompress the second channel report to obtain the recovered channel information. This can guarantee or improve the performance of the AI model when the network device performs different resource configurations to the terminal device, and further guarantee the performance of CSI feedback.

In conjunction with the second aspect, in some implementations of the second aspect, the method further includes: receiving fourth indication information, which indicates the configuration information of the second channel report, the configuration information of the second channel report including the maximum feedback overhead of the second channel report. Based on the above technical solution, the terminal device matches the configuration information of the channel report sent by the network device in real time and selects the most suitable AI model (e.g., a third AI model), which can guarantee or improve the performance of the AI model when the network device performs different resource configurations to the terminal device, and further guarantee the performance of CSI feedback.

In conjunction with the second aspect, in some implementations of the second aspect, the method further includes: determining the information of the second channel report based on the second resource configuration. Based on the above technical solution, the performance of the AI model can be guaranteed or improved when the network device performs different resource configurations to the terminal device, and the performance of CSI feedback can be further guaranteed.

In conjunction with the second aspect, in certain implementations of the second aspect, the information reported by the second channel includes at least one of the following: the number of resources corresponding to one or more measurement resources, the measurement results corresponding to one or more measurement resources, the time-domain information corresponding to one or more measurement resources, the frequency-domain information corresponding to one or more measurement resources, the spatial-domain information corresponding to one or more measurement resources, and the number of bits corresponding to each resource information in one or more measurement resources. Based on the above technical solution, the performance of the AI model can be guaranteed or improved when the network device performs different resource configurations to the terminal device, and the performance of CSI feedback can be further guaranteed.

Thirdly, a communication method is provided, which can be executed by a second device. The second device may include at least one of a network device or an AI entity on the network device side. The AI entity on the network device side can be the network device itself or an AI entity serving the network device (it can also be replaced by intelligent network elements, etc., without limitation), such as a radio access network (RAN), an intelligent controller (RIC), operation administration and maintenance (OAM), a server, etc., such as a cloud server. Here, an intelligent network element can also be understood as a network device with AI functionality, which can be applied in O-RAN architecture or non-O-RAN architecture, without limitation.

The method includes: determining a first resource configuration, the first resource configuration being one of at least one resource configuration supported by a first AI model, the first AI model being used to process channel information measured based on the first resource configuration; and sending first indication information, the first indication information indicating the first resource configuration.

It should be understood that the beneficial effects of the third aspect and any of its implementations can be referenced by the beneficial effects of the first aspect and any of its implementations.

In conjunction with the third aspect, in some implementations of the third aspect, the resource configuration includes: configuration type, offset of adjacent resources or number of transmissions, and the configuration type includes at least one of the following: periodic configuration, semi-static configuration or non-periodic configuration.

In conjunction with the third aspect, in some implementations of the third aspect, the method further includes: obtaining first information associated with the at least one resource configuration.

In conjunction with the third aspect, in some implementations of the third aspect, the first information includes the at least one resource configuration.

In conjunction with the third aspect, in some implementations of the third aspect, the first information includes the identification information of the first AI model.

In conjunction with the third aspect, in some implementations of the third aspect, the method further includes: obtaining second indication information, which indicates the correspondence between the identification information of the first AI model and the at least one resource configuration.

In conjunction with the third aspect, in some implementations of the third aspect, the second instruction information also indicates the correspondence between the identification information of the second AI model and at least one resource configuration supported by the second AI model, the second AI model being different from the first AI model.

In conjunction with the third aspect, in some implementations of the third aspect, receiving the second instruction information includes: obtaining second information, which includes the second instruction information, and the second information includes any one of the following: model-related information, registration request information, and terminal device capability information.

In conjunction with the third aspect, in some implementations of the third aspect, the first information includes functional information of the first AI model, which includes any one of the following:

Information on time-domain CSI prediction followed by compression function

Spatial and frequency domain CSI compression function information,

Spatial, frequency, and time domain CSI compression function information

Time-domain CSI prediction function information

Beam temporal prediction function information.

In conjunction with the third aspect, some implementations of the third aspect include: acquiring a second channel report and third indication information, wherein the third indication information indicates that the second channel report is associated with a third AI model, and the second channel report is generated by processing the channel information obtained from the second resource configuration measurement based on the third AI model.

In conjunction with the third aspect, in some implementations of the third aspect, the method further includes: sending fourth indication information, the fourth indication information indicating configuration information of the second channel report, the configuration information of the second channel report including the maximum feedback overhead of the second channel report.

In conjunction with the third aspect, in certain implementations of the third aspect, the information reported by the second channel includes at least one of the following:

The number of resources corresponding to one or more measurement resources

Measurement results corresponding to one or more measurement resources

Temporal information corresponding to one or more measurement resources

Frequency domain information corresponding to one or more measurement resources

Spatial information corresponding to one or more measurement resources

The number of bits corresponding to each resource information in one or more measurement resources.

Fourthly, a communication method is provided, which can be executed by a second device, the second device including at least one of a network device or an AI entity on the network device side. The AI entity on the network device side can be the network device itself or an AI entity serving the network device, such as a radio access network (RAN), a RAN intelligent controller (RIC), or an operation administration and maintenance (OAM) system.

The method includes: acquiring a second channel report and third indication information, wherein the third indication information indicates that the second channel report is associated with a third AI model, and the second channel report is generated by processing channel information obtained from second resource configuration measurements based on the third AI model.

It should be understood that the beneficial effects of the fourth aspect and any of its implementations can be referenced from the beneficial effects of the second aspect and any of its implementations.

In conjunction with the fourth aspect, in some implementations of the fourth aspect, the method further includes: sending fourth indication information, which indicates configuration information of the second channel report, the configuration information of the second channel report including the maximum feedback overhead of the second channel report.

In conjunction with the fourth aspect, in certain implementations of the fourth aspect, the information reported by the second channel includes at least one of the following:

The number of resources corresponding to one or more measurement resources

Measurement results corresponding to one or more measurement resources

Temporal information corresponding to one or more measurement resources

Frequency domain information corresponding to one or more measurement resources

Spatial information corresponding to one or more measurement resources

The number of bits corresponding to each resource information in one or more measurement resources.

Fifthly, a communication device is provided, which may be a first device, or a device or module for performing the functions of the first device.

One possible implementation is that the communication device may include modules or units corresponding to the methods/operations/steps/actions described in the first aspect, which may be hardware circuits, software, or a combination of hardware circuits and software.

The first device mentioned above can be a terminal device or an AI entity on the terminal device side, and there is no limitation thereto.

In a sixth aspect, a communication device is provided, which may be a second device, or a device or module for performing the functions of the second device.

One possible implementation is that the communication device may include modules or units corresponding to the methods/operations/steps/actions described in the second aspect, which may be hardware circuits, software, or a combination of hardware circuits and software.

The second device mentioned above can be a network device or an AI entity on the network device side.

A seventh aspect provides a communication device including a processor configured to, by executing a computer program or instructions, or by logic circuitry, cause the communication device to perform the method described in the first aspect and any possible manner of the first aspect; or cause the communication device to perform the method described in the second aspect and any possible manner of the second aspect; or cause the communication device to perform the method described in the third aspect and any possible manner of the third aspect; or cause the communication device to perform the method described in the fourth aspect and any possible manner of the fourth aspect.

In one possible implementation, the communication device also includes a memory for storing the computer program or instructions.

In one possible implementation, the communication device also includes a communication interface for inputting and/or outputting signals.

Eighthly, a communication device is provided, including logic circuitry and an input/output interface for inputting and/or outputting signals. The input/output interface is configured to perform the method described in the first aspect and any possible mode of the first aspect; or the logic circuitry is configured to perform the method described in the second aspect and any possible mode of the second aspect; or the logic circuitry is configured to perform the method described in the third aspect and any possible mode of the third aspect; or the logic circuitry is configured to perform the method described in the fourth aspect and any possible mode of the fourth aspect.

A ninth aspect provides a computer-readable storage medium storing a computer program or instructions that, when executed on a computer, cause the method described in the first aspect and any possible manner of the first aspect to be performed; or cause the method described in the second aspect and any possible manner of the second aspect to be performed; or cause the method described in the third aspect and any possible manner of the third aspect to be performed; or cause the method described in the fourth aspect and any possible manner of the fourth aspect to be performed.

In a tenth aspect, a computer program product is provided, comprising instructions that, when executed on a computer, cause the method described in the first aspect and any possible mode of the first aspect to be executed; or cause the method described in the second aspect and any possible mode of the second aspect to be executed; or cause the method described in the third aspect and any possible mode of the third aspect to be executed; or cause the method described in the fourth aspect and any possible mode of the fourth aspect to be executed.

Eleventhly, a chip or chip system is provided, comprising: one or more processors configured to execute computer programs or instructions in the memory, such that the chip or chip system implements the methods of the first aspect and any possible implementation thereof; or, such that the chip or chip system implements the methods of the second aspect and any possible implementation thereof; or, such that the chip or chip system implements the methods of the third aspect and any possible implementation thereof; or, such that the chip or chip system implements the methods of the fourth aspect and any possible implementation thereof.

For a description of the beneficial effects of any of the fifth to eleventh aspects, please refer to the description of the beneficial effects of the first to fourth aspects, which will not be repeated here.

Attached Figure Description

Figure 1 is a schematic diagram of an application framework applicable to an embodiment of this application.

Figure 2 is a schematic diagram of another application framework applicable to the embodiments of this application.

Figure 3 is a schematic diagram of a communication system applicable to an embodiment of this application.

Figure 4 is a schematic diagram of another communication system applicable to the embodiments of this application.

Figure 5 is a schematic diagram of the relationship between the encoder and the decoder.

Figure 6 is a schematic flowchart of a communication method 600 provided in an embodiment of this application.

Figure 7 is a schematic flowchart of a communication method 700 provided in an embodiment of this application.

Figure 8 is a schematic flowchart of a communication method 800 provided in an embodiment of this application.

Figure 9 is a schematic flowchart of a communication method 900 provided in an embodiment of this application.

Figure 10 is a schematic flowchart of a communication method 1000 provided in an embodiment of this application.

Figure 11 is a schematic flowchart of a communication method 1100 provided in an embodiment of this application.

Figure 12 is a schematic flowchart of another communication method 1200 according to an embodiment of this application.

Figure 13 is a schematic flowchart of another communication method 1300 according to an embodiment of this application.

Figure 14 is a schematic flowchart of another communication method 1400 according to an embodiment of this application.

Figure 15 is a schematic block diagram of a communication device according to an embodiment of this application.

Figure 16 is another schematic block diagram of a communication device according to an embodiment of this application.

Detailed Implementation

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

To facilitate understanding of the embodiments of this application, the following points will be explained first.

I. In this application, unless otherwise stated, "multiple" means two or more.

II. In the various embodiments of this application, unless otherwise specified or in case of logical conflict, the terms and/or descriptions between different embodiments are consistent and can be referenced by each other. The technical features in different embodiments can be combined to form new embodiments according to their inherent logical relationship.

III. The various numerical designations used in this application are merely for descriptive convenience and are not intended to limit the scope of protection of this application. The magnitude of the serial numbers used in this application does not imply a sequential order of execution; the execution order of each process should be determined by its function and internal logic. For example, the terms "first," "second," "third," "fourth," and other various terminology (if present) in the specification, claims, and drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. Such data can be interchanged where appropriate so that the embodiments described herein can be implemented in a sequence other than that illustrated or described herein.

Furthermore, any embodiment or design described in this application as "exemplary" or "for example" should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of terms such as "exemplary" or "for example" is intended to present the relevant concepts in a concrete manner for ease of understanding.

IV. The terms “comprising” and “having” and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are expressly listed, but may include other steps or units that are not expressly listed or that are inherent to such process, method, product or device.

V. In this application, "for indicating" can be understood as "enabling". "Enabling" can include direct enabling and indirect enabling. When describing information for enabling A, it can include whether the information directly enables A or indirectly enables A, but it does not mean that the information necessarily carries A.

The information that enables the information is called the information to be enabled. In the specific implementation process, there are many ways to enable the information to be enabled, such as, but not limited to, directly enabling the information to be enabled, such as the information to be enabled itself or its index. It can also be indirectly enabled by enabling other information, where there is a relationship between the other information and the information to be enabled. It can also enable only a part of the information to be enabled, while the other parts are known or pre-agreed upon. For example, enabling specific information can be achieved by using a pre-agreed (e.g., protocol-defined) arrangement of various pieces of information, thereby reducing enabling overhead to some extent. Simultaneously, common parts of various pieces of information can be identified and enabled uniformly to reduce the enabling overhead caused by individually enabling the same information.

VI. In this application, "pre-configuration" may include pre-defined terms, such as protocol definitions. These "pre-defined terms" can be implemented by pre-storing corresponding codes, tables, or other means of indicating relevant information in the device (e.g., including various network elements). This application does not limit the specific implementation method.

VII. The term "storage" or "preservation" in this application can refer to storage in one or more memory devices. These memory devices can be separately configured or integrated into an encoder or decoder, processor, or communication device. Alternatively, some memory devices can be separately configured, while others can be integrated into the decoder, processor, or communication device. The type of memory can be any form of storage medium, and this is not limited.

8. The “protocol” involved in this application may refer to standard protocols in the field of communications, such as fourth-generation (4G) network protocols, fifth-generation (5G) network protocols, new radio (NR) protocols, 5.5G network protocols, future network protocols, and related protocols applied in future communication systems. This application does not limit the scope of the term.

9. The arrows or boxes indicated by dashed lines in the schematic diagrams in the accompanying drawings of this application represent optional steps or optional modules.

10. In this application, unless otherwise stated, "/" indicates that the objects before and after are in an "or" relationship. For example, A/B can mean A or B. In this application, "and/or" is merely a description of the relationship between the related objects, indicating that there can be three relationships. For example, A and/or B can mean: A exists alone, A and B exist simultaneously, and B exists alone. A and B can be singular or plural.

XI. In this application, the indication includes direct indication (also known as explicit indication) and implicit indication. Direct indication information A means including information A. Implicit indication information A means indicating information A through the correspondence between information A and information B, and through direct indication information B. The correspondence between information A and information B can be predefined, pre-stored, pre-burned, or pre-configured.

12. In this application, information C is used to determine information D, including both when information D is determined solely based on information C and when it is determined based on information C and other information. Furthermore, information C can also be used to determine information D indirectly, for example, when information D is determined based on information E, and information E is determined based on information C.

Thirteen, in this application, "device A sends information A to device B" can be understood as device B being the destination of information A or an intermediate network element in the transmission path between the destination and the destination, which may include sending information to device B directly or indirectly.

14. In this application, "device B receives information A from device A" can be understood as device A being the source of information A or an intermediate network element in the transmission path between device B and device A. This can include receiving information directly or indirectly from device A. Information may undergo necessary processing, such as format changes, between the source and destination ends, but the destination end can understand the valid information from the source end. Similar expressions in this application can be interpreted similarly and will not be elaborated further here.

First, the communication system to which the embodiments of this application are applicable will be described.

The technical solutions provided in this application can be applied to various communication systems, such as 5G or 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 such as future mobile 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.

In a communication system, a device can send signals to or receive signals from another device. Signals may include information, signaling, or data. A device can also be replaced by an entity, network entity, equipment, communication device, communication module, node, communication node, etc. This application describes a device as an example. For instance, a communication system may 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, and/or, the terminal device can send sidelink signals to another terminal device, and/or, the network device can send signals to another network device.

In the embodiments of this application, the terminal device may also be referred to as user equipment (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 device.

Terminal devices can be devices that provide voice/data, such as handheld devices and in-vehicle devices with wireless connectivity. Currently, some 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, and wireless terminals in transportation safety. The embodiments of this application do not limit the scope to wireless terminals in smart cities, 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 terminal devices in future evolved public land mobile networks (PLMNs).

As an example, and not a limitation, 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; they achieve powerful functions through software support, data interaction, and cloud interaction. Broadly defined, wearable smart devices include those that are feature-rich, large in size, and can achieve complete or partial functionality without relying on a smartphone, such as smartwatches or smart glasses, as well as those that focus on a specific application function and require the use of other devices such as smartphones, such as various smart bracelets and smart jewelry for vital sign monitoring.

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.

The network device in this application embodiment may include a device for communicating with a terminal device. This network device may include an access network device or a radio access network device, such as a base station. The network device in this application embodiment may also include a radio access network (RAN) node (or device) for connecting the terminal device to a wireless network.

A base station can broadly encompass, or be replaced by, various names including: 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, and access point. The network includes points (APs), transmission nodes, transceiver nodes, baseband units (BBUs), remote radio units (RRUs), active antenna units (AAUs), remote radio heads (RRHs), central units (CUs), distributed units (DUs), radio units (RUs), positioning nodes, and RAN intelligent controllers (RICs).

A base station can also be a macro base station, micro base station, relay node, donor node, or similar entity, or a combination 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 and equipment performing base station functions in D2D, V2X, and M2M communications, network-side equipment in future networks, or equipment performing base station functions in future communication systems. A base station can support networks using the same or different access technologies.

Optionally, RAN nodes can also be servers, wearable devices, vehicles, or in-vehicle equipment. For example, the access network equipment in vehicle-to-everything (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 network devices.

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.

In some deployments, network devices can be devices that include CUs or DUs, or devices that include both CUs and DUs, or devices with control plane CU nodes (central unit-control plane (CU-CP)) and user plane CU nodes (central unit-user plane (CU-UP)) and DU nodes. For example, network devices include gNB-CU-CP, gNB-CU-UP, and gNB-DU.

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.

RAN nodes can support one or more types of fronthaul interfaces, and different fronthaul interfaces correspond to 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 DU and RU is a different interface, relative to CPRI, some baseband functions for downlink and/or uplink, 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 DU to RU; and for uplink, digital beamforming (BF), or one or more of fast fourier transform (FFT)/cyclic prefix removal (CP) are moved from DU to RU.

One possible implementation is that the interface can be an enhanced common public radio interface (eCPRI). In 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.

Taking eCPRI Cat A as an example, for downlink transmission, the DU is configured to implement one or more functions before and after the layer mapping (i.e., coding, rate matching, scrambling, modulation, layer mapping). Other functions after the layer mapping (e.g., resource element (RE) mapping, digital beamforming (BF), or one or more of inverse fast Fourier transform (IFFT)/adding a cyclic prefix (CP)) are moved to the RU for implementation. For uplink transmission, the de-RE mapping is used as the dividing line. The 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 one or more functions from de-RE mapping). Other functions following de-mapping (e.g., one or more functions from digital BF or fast Fourier transform (FFT)/CP removal) are moved to the RU. It is understood that descriptions of the functions of the DU and RU corresponding to various types of eCPRI can be found in the eCPRI protocol and will not be elaborated upon here.

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.

In different communication 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) system, CU can also be called O-CU (open CU), DU can also be called O-DU, CU-CP can also be called O-CU-CP, CU-UP can also be called O-CU-UP, and RU can also be called 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.

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.

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, software functions running on dedicated hardware, software functions running on 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 terminal devices and network devices.

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

To support AI technology in wireless networks, AI nodes (also known as AI entities) may be introduced into the network.

Optionally, the AI entity can be deployed in one or more of the following locations within the communication system: access network devices, terminal devices, or core network devices, or the AI entity can be deployed independently, for example, in a location other than any of the aforementioned devices, such as the host or cloud server of an OTT system. The AI entity can communicate with other devices in the communication system, which can be one or more of the following: network devices, terminal devices, or network elements of the core network. Based on the object served by the AI entity, the AI entity can include an AI entity on the network device side, an AI entity on the terminal device side, or an AI entity on the core network side.

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

It can also be understood that AI entities can be independent devices, or they can be integrated into the same device to achieve different functions. Alternatively, they can be network components 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 entities.

AI entities can be AI network elements or AI modules. AI entities are used to implement corresponding AI functions. AI modules deployed in different network elements can be the same or different. Depending on the different parameter configurations, the AI model within an AI entity can achieve different functions. The AI model within an AI entity 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 biases in the activation function), input parameters (e.g., the type and/or dimension of the input parameters), or output parameters (e.g., the type and/or dimension of the output parameters). The biases in the activation function can also be referred to as the biases of the neural network.

An AI entity can have one or more models. The learning, training, or inference processes of different models can be deployed in different entities or devices, or they can be deployed in the same entity or device.

Figure 1 is a schematic diagram of an application framework applicable to an embodiment of this application. As shown in Figure 1, devices are connected via interfaces (e.g., NG, Xn) or over-the-air interfaces. These device nodes, such as core network devices, access network nodes (RAN nodes), terminals, or one or more devices in OAM, are equipped with one or more AI modules (only one is shown in Figure 1 for clarity). An access network node can be a single RAN node or can include multiple RAN nodes, for example, including CU and DU. CU and/or DU can also be equipped with one or more AI modules. Optionally, a CU can also be split into CU-CP and CU-UP. One or more AI models are configured in CU-CP and/or CU-UP.

This AI module is used to implement corresponding AI functions. AI modules deployed in different devices can be the same or different. Depending on the parameter configuration, the AI module can achieve 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 biases in the activation function), input parameters (e.g., the type and/or dimension of the input parameters), or output parameters (e.g., the type and/or dimension of the output parameters). The biases in the activation function can also be referred to as the biases of the neural network.

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.

Figure 2 is a schematic diagram of another application framework applicable to the embodiments of this application. As shown in Figure 2, the communication system includes a Resource Interchange (RIC). For example, the RIC can be the AI module in the RAN node shown in Figure 1, used to implement AI-related functions. RICs include near-real-time RICs (near-RT RICs) and non-real-time RICs (non-RT RICs). Non-real-time RICs mainly process non-real-time information, such as data that is not sensitive to latency, with latency in the order of seconds. Real-time RICs mainly process near-real-time information, such as data that is relatively sensitive to latency, with latency in the order of tens of milliseconds.

Near real-time RICs are used for model training and inference. For example, they are used to train AI models and then use those models for inference. Near real-time RICs 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, 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, near real-time RIC submits inference results to DU, and DU sends them to RU.

Non-real-time RICs are also used for model training and inference. For example, they are used to train AI models and then use those models for inference. Non-real-time RICs 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, 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, a non-real-time RIC can submit inference results to DU, which in turn can send them to RU.

Near real-time RICs and non-real-time RICs can also be configured as separate devices. Alternatively, near real-time RICs and non-real-time RICs can also be part of other devices. For example, near real-time RICs can be configured in RAN nodes (e.g., CU, DU), while non-real-time RICs can be configured in OAM, cloud servers, core network devices, or other devices.

Figure 3 is a schematic diagram of a communication system applicable to an embodiment of this application. As shown in Figure 3, the communication system may include at least one network device, such as network device 110; the communication system 100 may also include at least one terminal device, such as terminal device 120 and terminal device 130. Network device 110 and terminal devices (such as terminal devices 120 and 130) can communicate via a wireless link. The communication devices in this communication system, for example, network device 110 and terminal device 120, can communicate via multi-antenna technology.

Figure 4 is a schematic diagram of another communication system applicable to the embodiments of this application. Compared with the communication system described in Figure 3, the communication system shown in Figure 4 further includes an AI device 140, which is used to perform AI-related operations, such as building a training dataset or training an AI model. The AI device 140 is the aforementioned AI node or AI entity.

In one possible implementation, network device 110 sends data related to the training of the AI model to AI device 140, whereby AI device 140 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 the terminal device. AI device 140 sends the results of operations related to the AI model to network device 110, which then forwards them to the terminal device. 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 is deployed on network device 110, and another portion is deployed on the terminal device. Alternatively, the trained AI model is deployed on network device 110. Or, the trained AI model is deployed on the terminal device.

It should be understood that Figure 4 only illustrates the example of AI device 140 being directly connected to network device 110. In other scenarios, AI device 140 can also be connected to terminal devices; AI device 140 can also be connected to both network device 110 and terminal devices simultaneously; AI device 140 can also be connected to network device 110 through third-party devices, etc. Therefore, this application does not limit the connection relationship between AI device and other devices.

The AI device 140 can also be installed as a module in network devices and/or terminal devices, for example, in network device 110 or terminal device as shown in FIG3.

It should be noted that Figures 3 and 4 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 3 and 4. In practical applications, the communication system may include multiple network devices (such as network device 110 and network device 150 (not shown in Figure 3)) and multiple terminal devices. Therefore, this application does not limit the number of network devices and terminal devices included in the communication system.

Next, a brief description of some of the technical concepts involved in this application will be given.

(1) AI Model:

An AI model is an algorithm or computer program that enables AI functionality. It represents the mapping relationship between the model's input and output. An AI model can be understood as a function model that maps an input of a certain dimension to an output of a certain dimension, and its parameters are obtained through machine learning training. For example, f(x) = a* ^x² + b is a quadratic function model, which can be considered an AI model. a and b correspond to the parameters of the AI model, which can be obtained through machine learning training. An AI model can also be called a model, an AI function, or a function. One AI function can correspond to one or more AI models.

AI models can be neural networks, linear regression models, decision tree models, support vector machines (SVM), Bayesian networks, Q-learning models, or other machine learning (ML) models.

(2) Two-ended model:

A two-sided model, also known as a bilateral model, collaborative model, dual model, or two-side model, refers to a model composed of multiple sub-models. These sub-models need to be mutually compatible and can be deployed on different nodes.

This application embodiment relates to an encoder for compressing CSI and a decoder for recovering CSI. The encoder and decoder are used in conjunction, and can be understood as paired AI models. An encoder may include one or more AI models, and the decoder matched with the encoder also includes one or more AI models. The number of AI models included in the matched encoder and decoder are the same and correspond one-to-one. The encoder may also include a quantization module, which can be used to quantize the output of the AI model in the encoder. The decoder may include an inverse quantization module, which can be used to inverse quantize the feedback information of the received channel information to obtain the input of the AI model in the decoder. Inverse quantization can also be replaced by dequantization.

In one possible design, a set of matched encoders and decoders can be two parts of the same autoencoder (AE). An AE model where the encoder and decoder are deployed on different nodes is a typical bilateral model. In other AE models, the encoder and decoder are usually co-trained and used in combination. An autoencoder is an unsupervised learning neural network that uses input data as labeled data; therefore, it can also be understood as a self-supervised learning neural network. Autoencoders can be used for data compression and reconstruction. For example, the encoder in an autoencoder can compress (encode) data A to obtain data B; the decoder in an autoencoder can decompress (decode) data B to recover data A. Alternatively, the decoder can be understood as the inverse operation of the encoder. A description of the encoder and decoder can be found in Figure 5.

Figure 5 is a schematic diagram of the relationship between the encoder and the decoder. As shown in Figure 5, the encoder processes the input V to obtain the processed result z, and the decoder can decode the encoder's output z into the desired output V'.

The self-encoding model in this application embodiment may include an encoder deployed on the terminal device side and a decoder deployed on the network device side, or an encoder deployed on the terminal device side and a decoder deployed on another terminal device side, or an encoder deployed on the network device side and a decoder deployed on another network device side.

(3) Neural Network (NN):

Neural networks are a specific implementation of AI or machine learning. According to the general approximation theorem, neural networks can theoretically approximate any continuous function, thus enabling them to learn arbitrary mappings.

A neural network can be composed of neural units, which can be defined as computational units that take x, _s , and an intercept of 1 as inputs. A neural network is a network formed by connecting many of these individual neural units together; that is, the output of one neural unit can be the input of another. The input of each neural unit can be connected to the local receptive field of the previous layer to extract features from that local receptive field, which can be a region composed of several neural units.

Taking neural networks as an example, the AI model involved in this application can be a deep neural network (DNN). Depending on the construction method of the network, DNN can include feedforward neural networks (FNN), convolutional neural networks (CNN), and recurrent neural networks (RNN), etc.

CNNs are neural networks specifically designed to process data with a grid-like structure. For example, time-series data (discrete sampling along the time axis) and image data (two-dimensional discrete sampling) can both be considered grid-like data. CNNs do not use all the input information at once for computation; instead, they use a fixed-size window to extract a portion of the information for convolution operations, which significantly reduces the computational cost of model parameters. Furthermore, depending on the type of information extracted by the window (such as people and objects in an image representing different types of information), each window can use different convolution kernels, allowing CNNs to better extract features from the input data.

Recurrent Neural Networks (RNNs) are a type of distributed neural network (DNN) that utilizes feedback time-series information. Their input includes the current input value and their own output value from the previous time step. RNNs are well-suited for acquiring temporally correlated sequence features, and are particularly applicable to applications such as speech recognition and channel encoding/decoding.

The characteristic of FNN networks is that neurons in adjacent layers are completely connected to each other, which makes FNNs typically require a large amount of storage space and result in high computational complexity.

The FNN, CNN, and RNN mentioned above are all constructed based on neurons. As mentioned earlier, each neuron performs a weighted summation operation on its input values, and the result of the weighted summation is used to generate the output through a nonlinear function. The weights of the weighted summation operation of neurons in a neural network and the nonlinear function are called the parameters of the neural network. The parameters of all neurons in a neural network constitute the parameters of that neural network.

(4) Dataset:

A dataset refers to the data used for model training, validation, and testing in machine learning. The quantity and quality of the data will affect the effectiveness of machine learning.

In the field of machine learning, ground truth usually refers to data that is considered accurate or real.

Training datasets are used to train AI models. A training dataset can include the input to the AI model, or it can include both the input and the target output of the AI model. Specifically, a training dataset includes one or more training data points, which can include training samples input to the AI model or the target output of the AI model. The target output can also be referred to as a label, sample label, or labeled sample. The label is the ground truth value.

In the field of communications, training datasets can include simulation data collected through simulation platforms, experimental data collected from experimental scenarios, or measured data collected in actual communication networks. Because the geographical environment and channel conditions where the data is generated vary—for example, indoor/outdoor conditions, movement speed, frequency bands, or antenna configurations—the collected data can be categorized during acquisition. For instance, data with the same channel propagation environment and antenna configuration can be grouped together.

Model training essentially involves learning certain features from training data. In training AI models (such as neural network models), the goal is to make the model's output as close as possible to the desired predicted value. This is achieved by comparing the network's current predictions with the target value and updating the weight vector of each layer based on the difference. (Of course, there's usually an initialization process before the first update, where parameters are pre-configured for each layer.) For example, if the network's prediction is too high, the weight vector is adjusted to predict a lower value. This adjustment continues until the AI model can predict the target value or a value very close to it. Therefore, it's necessary to predefine "how to compare the difference between the predicted and target values," which is the loss function or objective function. These are important equations used to measure the difference between the predicted and target values. Taking the loss function as an example, a higher output value (loss) indicates a greater difference. Therefore, training the AI model becomes a process of minimizing this loss, making the loss function value less than a threshold, or making the loss function value meet the target requirements. For example, if the AI model is a neural network, adjusting the model parameters of the neural network includes adjusting at least one of the following parameters: the number of layers of the neural network, the width, the weights of the neurons, or the parameters in the activation function of the neurons.

Inference data can be used as input to a trained AI model for inference. During the model's inference process, the inference data is input into the AI model, and the corresponding output, which is the inference result, is obtained.

(5) AI model design:

The design of an AI model mainly includes the data collection phase (e.g., collecting training data and/or inference data), the model training phase, and the model inference phase. It can also further include the application of the inference results.

The training processes of different models can be deployed on different devices or nodes, or on the same device or node. Similarly, the inference processes of different models can be deployed on different devices or nodes, or on the same device or node. Taking a terminal device completing the model training phase as an example, after training its corresponding encoder and decoder, the terminal device sends the decoder's model parameters to the network device. Similarly, taking a network device completing the model training phase as an example, after training its corresponding encoder and decoder, the network device can send the encoder's model parameters to the terminal device and the decoder's model parameters to the network device. Then, the model inference phase corresponding to the encoder is performed on the terminal device, and the model inference phase corresponding to the decoder is performed on the network device.

The model parameters can include one or more of the following: model structure parameters (e.g., number of layers, and/or weights), model input parameters (e.g., input dimension, number of input ports), or model output parameters (e.g., output dimension, number of output ports). The input dimension refers to the size of an input data set; for example, if the input data is a sequence, the corresponding input dimension indicates the length of the sequence. The number of input ports refers to the quantity of input data. Similarly, the output dimension refers to the size of an output data set; for example, if the output data is a sequence, the corresponding output dimension indicates the length of the sequence. The number of output ports refers to the quantity of output data.

(6) Channel information:

In communication systems (e.g., LTE or NR systems), network devices determine one or more configurations, such as downlink data channel resources, MCS (Multi-Channel System), and precoding, for scheduling terminal devices based on channel information. Channel information, also known as channel state information (CSI) or channel environment information, is information that reflects channel characteristics and quality.

Channel information measurement refers to the process by which the receiver deciphers the channel information based on a reference signal transmitted by the transmitter; that is, it involves estimating the channel information using channel estimation methods. For example, the reference signal may include one or more of the following: channel state information reference signal (CSI-RS), synchronizing signal/physical broadcast channel block (SSB), sounding reference signal (SRS), or demodulation reference signal (DMRS). One or more of CSI-RS, SSB, and DMRS can be used to measure downlink channel information. SRS and/or DMRS can be used to measure uplink channel information.

Channel information can be determined based on channel measurement results of a reference signal. Alternatively, channel information can be the channel measurement results of a reference signal. In the embodiments of this application, the channel measurement results of the reference signal or the channel measurement results can also be replaced by channel information.

Taking FDD communication as an example, in FDD communication, because uplink and downlink channels lack reciprocity or cannot guarantee reciprocity, network devices need to obtain downlink CSI through uplink feedback from terminal devices. Network devices typically send a downlink reference signal to the terminal device, which receives this signal. Since the terminal device knows the transmission information of the downlink reference signal, it can perform channel measurement and interference measurement based on the received signal to estimate the downlink channel traversed by the downlink reference signal. The terminal device then generates the downlink CSI based on this measurement and the resulting downlink channel matrix. The terminal device generates a CSI report according to a predefined protocol method or a network device configuration method and feeds it back to the network device so that it can obtain the downlink CSI.

In this application, the meaning of CSI is broader than that of CSI in traditional schemes. It is not limited to channel quality indicator (CQI), precoding matrix indicator (PMI), rank indicator (RI), or CSI-RS resource indicator (CRI). It can also be one or more of the following: channel response information (such as channel response matrix, frequency domain channel response information, time domain channel response information), weight information corresponding to the channel response, reference signal receiving power (RSRP), or signal to interference plus noise ratio (SINR).

In this context, RI indicates the recommended number of downlink transmission layers for the receiving end of the reference signal, such as a terminal device; CQI indicates the modulation and coding schemes supported by the current channel conditions for the receiving end of the reference signal, such as a terminal device; and PMI indicates the recommended precoding for the receiving end of the reference signal, such as a terminal device. The number of precoding layers indicated by PMI corresponds to RI.

As mentioned earlier, channel information can be obtained by measuring the reference signal. Compressing and/or quantizing this channel information yields feedback information. This feedback information can be reported via a channel information report. Therefore, feedback information can also be referred to as a channel report. In the embodiments of this application, a channel report may include at least one sub-channel report.

The channel information can be recovered by decompressing and/or inverting the feedback information. Decompression can also be understood as recovery or reconstruction, which will not be elaborated further below.

Feedback information can also be called channel information feedback information, CSI feedback information, CSI feedback information, compressed information, compressed channel information, compressed CSI information, compressed channel information, or compressed CSI, etc.

The recovered channel information can also be called CSI recovery information.

As the antenna array size of MIMO systems continues to increase, the number of supported antenna ports also increases, leading to a growth in the dimensionality of the corresponding channel matrix and precoding matrix. To enable terminal devices to estimate (measure) the downlink channel, the overhead of network devices transmitting reference signals increases. Simultaneously, the error in approximating large-scale channel and precoding matrices using a finite number of predefined codewords increases. One method to improve channel recovery accuracy is to increase the number of codewords in the codebook; however, this simultaneously increases the overhead of CSI feedback (including the corresponding codeword number and one or more weighting coefficients), thereby reducing available resources for data transmission and causing system capacity loss.

Introducing AI technology into wireless communication networks has led to a CSI feedback method based on AI models, known as AI-CSI feedback. Terminal devices use AI models to compress and feed back CSI, while network devices use AI models to reconstruct the compressed CSI. AI-based CSI feedback transmits a sequence (such as a bit sequence), resulting in lower overhead compared to traditional CSI feedback. Furthermore, AI models possess stronger nonlinear feature extraction capabilities, enabling more effective compression and representation of channel information and more efficient channel reconstruction based on feedback information compared to traditional methods.

CSI feedback can be implemented based on AE's AI model. Taking Figure 5 as an example, the encoder in Figure 5 can be a CSI generator, and the decoder can be a CSI reconstructor. For example, the encoder can be deployed in the terminal device, and the decoder can be deployed in the network device.

The channel information V is used by the encoder to generate CSI feedback information z. The channel information is then reconstructed by the decoder to obtain the recovered channel information V’.

Channel information V can be obtained through channel information measurement. For example, channel information V may include the eigenvector matrix (a matrix composed of eigenvectors) of the downlink channel. The encoder processes the eigenvector matrix of the downlink channel to obtain CSI feedback information z. In other words, the compression and/or quantization operations on the eigenvector matrix based on the codebook in the correlation scheme are replaced by the operation of the encoder processing the eigenvector matrix to obtain CSI feedback information z. The decoder processes the CSI feedback information z to obtain the recovered channel information V'.

The training and inference processes of the AI model in the embodiments of this application will be further illustrated below.

Training data used to train an AI model includes training samples and sample labels. For example, training samples are channel information measured by the terminal device, and sample labels are the actual channel information, i.e., the ground truth CSI. If the encoder and decoder belong to the same autoencoder, the training data may only include training samples, or in other words, the training samples are the sample labels.

In the field of wireless communication, the true CSI can be a high-precision CSI.

The specific training process is as follows: The model training node uses an encoder to process channel information, i.e., training samples, to obtain CSI feedback information, and uses a decoder to process the feedback information to obtain the recovered channel information, i.e., the CSI recovered information. Then, the difference between the CSI recovered information and the corresponding sample label is calculated, i.e., the value of the loss function. The parameters of the encoder and decoder are updated according to the value of the loss function to minimize the difference between the recovered channel information and the corresponding sample label, i.e., minimizing the loss function. For example, the loss function can be the minimum mean square error (MSE) or cosine similarity. Repeating the above operations yields an encoder and decoder that meet the target requirements. The above model training node can be a terminal device, network device, or other network element with AI functionality in a communication system. The AI model can be implemented in hardware circuits, software, or a combination of both. Non-limiting examples of software include: program code, program, subroutine, instruction, instruction set, code, code segment, software module, application program, or software application, etc.

Currently, the configuration of reference signals sent by network devices to terminal devices is flexible and variable, which may cause performance fluctuations in AI models during channel information processing. For example, if the AI model on the terminal device side is an AI CSI compression model, it may lead to performance fluctuations in CSI feedback. Similarly, if the AI model on the terminal device side is an AI CSI prediction model, it may lead to performance fluctuations in CSI prediction.

Based on this, this application aims to provide a communication method that can guarantee or improve the performance of AI models when network devices perform different resource configurations to terminal devices.

The communication method provided in the embodiments of this application will be described in detail below with reference to the accompanying drawings. It should be noted in advance that Figures 6 to 11 are illustrated using the example of the first device as a terminal device and the second device as a network device. For ease of understanding, in the relevant content and description of Figures 6 to 11, the terminal device is used to replace the first device and the network device is used to replace the second device.

Figure 6 is a schematic flowchart of a communication method 600 provided in an embodiment of this application. As shown in Figure 6, the method includes at least the following steps.

S610, network devices determine the first resource allocation.

S620, the network device sends a first instruction message to the terminal device, and the terminal device receives the first instruction message accordingly.

Specifically, in steps S610 and S620, the first indication information indicates a first resource configuration, which is one of at least one resource configuration supported by the first AI model. After receiving the first resource configuration, the terminal device measures channel information based on the first resource configuration and processes the channel information according to the first AI model.

For example, when the first AI model is an AI CSI compression model, processing the channel information according to the first AI model includes: compressing the channel information according to the first AI model.

For example, when the first AI model is an AI CSI prediction followed by compression model, processing the channel information according to the first AI model includes: performing prediction processing on the channel information according to the first AI model, and then further compressing the channel information obtained from the prediction processing. The AI CSI prediction followed by compression model can also be called an AI CSI prediction plus compression model or an AI CSI prediction + compression model; it should be understood that this application does not impose any limitations on this.

Optionally, in the two examples described above, the method may further include: the terminal device determining a first channel report. This can be understood as the first channel report being determined by the terminal device after compressing the channel information obtained from the first resource configuration measurement based on the first AI model.

For example, when the first AI model is an AI CSI prediction or an AI beamforming time-domain prediction model, processing the channel information according to the first AI model includes: performing prediction processing on the channel information according to the first AI model. Optionally, in this case, the method may further include: the terminal device performing compression processing on the channel information obtained from the prediction processing according to a non-AI model.

In this embodiment, resource configuration includes configuration type, offset of adjacent resources, or number of transmissions, wherein the configuration type includes at least one of the following: periodic configuration, semi-static configuration, or aperiodic configuration. For example, resource configuration can be CSI-RS configuration or configuration of other reference signals.

For example, when CSI-RS is configured for periodic configuration, the network device configures the transmission period (e.g., every N slots) and offset (slot offset within the period) of the CSI-RS resources and informs the terminal device. The transmission period of the CSI-RS resources can be understood as the offset between adjacent CSI-RS resources.

For example, when CSI-RS is configured in a semi-static manner, the network device configures the transmission period (e.g., every N slots) and offset (time slot offset within the period) of the CSI-RS resource and informs the terminal device. The network device can activate or deactivate the transmission of the CSI-RS resource using configuration information such as the medium access control-control element (MAC-CE).

For example, when CSI-RS is configured for aperiodic configuration, the network device notifies the terminal device of each CSI-RS resource transmission via downlink control information (DCI) signaling. Furthermore, in one scenario, aperiodic CSI-RS configuration also supports configuring the transmission of one or more CSI-RS resources at a time. For instance, the network device configures this using a set of parameters [m, K], where m is the transmission period of the aperiodic CSI-RS resources (e.g., every m slots), and K is the number of CSI-RS resource transmissions. The transmission period of the CSI-RS resources can be understood as the offset between adjacent CSI-RS resources, and the number of CSI-RS resource transmissions can be understood as the number of CSI-RS resources; these details will not be elaborated further below.

It should be noted that the feedback from the channel report also supports the above three configuration types.

It should also be noted that in step S610, the network device determines the first resource configuration in several possible ways.

Method 1:

Alternatively, in one possible implementation, the network device obtains at least one resource configuration supported by the first AI model through protocol predefinition, and then the network device selects a first resource configuration from the at least one resource configuration.

Alternatively, in one possible implementation, the network device obtains not only at least one resource configuration supported by a first AI model through protocol predefinition, but also at least one resource configuration supported by a second AI model through protocol predefinition. Subsequently, the network device selects a first resource configuration from all the obtained resource configurations. It should be understood that the second AI model can be understood as one or more AI models different from the first AI model.

Method 2:

Optionally, in one possible implementation, before step S610, the method may further include: the terminal device sending first information to the network device, and correspondingly, the network device receiving the first information, wherein the first information is associated with at least one resource configuration.

For example, in one possible implementation, the first information includes at least one resource configuration supported by the first AI model.

Specifically, the terminal device sends first information to the network device, which includes at least one resource configuration supported by the first AI model on the terminal device side. That is to say, the terminal device reports at least one resource configuration supported by the first AI model to the network device.

For example, if there is only one AI model (e.g., AI model #1) on the terminal device side, the terminal device reports at least one resource configuration supported by AI model #1. For example, as shown in Table 1, AI model #1 supports the following resource configurations.

Table 1

As shown in Table 1, the AI model #1 on the terminal device side supports four resource configurations (Resource Configuration #1 to Resource Configuration #4). Resource Configuration #1 is a periodic CSI-RS configuration, Resource Configuration #2 is a semi-static CSI-RS configuration, and Resource Configurations #3 and #4 are aperiodic CSI-RS configurations. Specifically, the offset of the adjacent resources for the aperiodic CSI-RS configuration corresponding to Resource Configuration #3 is m1, and the number of CSI-RS transmissions is K1. The offset of the adjacent resources for the aperiodic CSI-RS configuration corresponding to Resource Configuration #4 is m2, and the number of CSI-RS transmissions is K2.

After receiving the above four resource configurations, the network device selects one resource configuration (i.e., the first resource configuration) from the four resource configurations and sends it to the terminal device.

For example, in one possible implementation, the first information also includes at least one resource configuration supported by the second AI model. The second AI model can be understood as one or more AI models different from the first AI model.

For example, when there are multiple AI models on the terminal device side, the terminal device can also report at least one resource configuration supported by all AI models. It should be noted that in this case, the terminal device reports multiple AI models in multiple installments. Assuming there are three AI models on the terminal device side, in one scenario, the terminal device reports the resource configurations supported by each AI model in three separate reports, with each report occurring at a different time. For instance, the terminal device reports the resource configuration supported by AI model #1 at the first time, AI model #2 at the second time, and AI model #3 at the third time. It should be understood that the first, second, and third times are different. Therefore, the network device can determine the different AI models and the resource configurations supported by each AI model based on the resource configurations received at different times. For example, the network device might determine that the resource configuration received at the first time is supported by AI model #1.

For example, in another scenario, the terminal device reports different resource configuration indications for each AI model; in other words, the AI models can report different resource configurations corresponding to different AI models through different indications. For instance, the terminal device reports the resource configuration supported by AI model #1 through indication #A, the terminal device reports the resource configuration supported by AI model #2 through indication #B, and the terminal device reports the resource configuration supported by AI model #3 through indication #C. It should be understood that indications #A, #B, and #C are different. Then, the network device can determine different AI models and the resource configurations supported by different AI models based on the different indications received. For example, the network device can determine that the resource configuration received through indication #A is supported by AI model #1.

It should be understood that the above examples are merely illustrative. The resource configuration of each AI model reported by the terminal device in multiple reports can also be associated with other information (other than the above-mentioned indication information), and this application does not impose any restrictions on this.

It should also be understood that the above time can be replaced with other time information, and this application does not impose any restrictions on this.

See Table 2. AI Models #1 to #3 support the following resource configurations respectively.

Table 2

As shown in Table 2, AI model #1 on the terminal device side supports four resource configurations (resource configuration #1-resource configuration #4), AI model #2 supports three resource configurations (resource configuration #1-resource configuration #3), and AI model #3 also supports three resource configurations (resource configuration #1-resource configuration #3). The relevant descriptions of the resource configurations for AI model #1 can be found above and will not be repeated here. There are no restrictions on the offset of adjacent resources and the number of times CSI-RS resources are sent for the aperiodic configuration corresponding to AI model #2. For AI model #3, the offset of adjacent resources for the aperiodic configuration CSI-RS is m3, and the number of times CSI-RS resources are sent is K3.

Furthermore, after receiving the resource configuration reported by the terminal device, the network device selects a resource configuration (i.e., the first resource configuration) from the resource configurations and sends it to the terminal device.

It should be noted that the AI models #1, #2, and #3 mentioned above are only for the purpose of illustrating that these are three different AI models and are not related to the identification information of the three AI models.

For example, in another possible implementation, the first information includes identification information of the first AI model. This identification information identifies the first AI model; for instance, it can be a number sequence identifier, an letter sequence identifier, or a dataset identifier, indicating that the first AI model was trained on a specific dataset. For example, a device may have multiple datasets, and the device can train multiple AI models based on these datasets. The dataset identifier can indicate the dataset used by a particular AI model. As an example, different datasets can be identified using different identifiers.

It should be noted that the identification information of the first AI model can be randomly generated or predefined, and it should be understood that this application does not impose any restrictions on this.

It should be understood that the identification information of the first AI model mentioned above is merely an example, and this application does not impose any limitations on it.

Specifically, in one possible implementation, the network device pre-defines the correspondence between the identification information of the first AI model (e.g., identifier 1) and at least one resource configuration through a protocol predefined structure, as shown in Table 3. In this embodiment, the terminal device sends the identification information of the first AI model (e.g., identifier 1) to the network device. Based on the received identifier 1 and the correspondence between the identification information of the first AI model and at least one resource configuration obtained through the protocol predefined structure, the network device determines at least one resource configuration supported by the first AI model corresponding to identifier 1, and selects one resource configuration (e.g., the first resource configuration) from the at least one resource configuration and sends it to the terminal device. The correspondence can also be referred to as an association relationship, etc., and is not limited here.

Table 3

As shown in Table 3, the AI model corresponding to identifier 1 supports four resource configurations (Resource Configuration #1 to Resource Configuration #4). Resource Configuration #1 is a periodic CSI-RS configuration, Resource Configuration #2 is a semi-static CSI-RS configuration, and Resource Configurations #3 and #4 are aperiodic CSI-RS configurations. Specifically, the offset of adjacent resources for the aperiodic CSI-RS configuration corresponding to Resource Configuration #3 is m4, and the number of CSI-RS transmissions is K4. The offset of adjacent resources for the aperiodic CSI-RS configuration corresponding to Resource Configuration #4 is m5, and the number of CSI-RS transmissions is K5.

It should be noted that, in another possible implementation, the network device can also pre-define the identification information of other AI models on the terminal device side and the correspondence between them and at least one resource configuration through protocol pre-definition. For example, the network device can also pre-define the identification information of a second AI model and the correspondence between it and at least one resource configuration supported by the second AI model. Here, the second AI model can be understood as one or more AI models different from the first AI model.

The identification information of the second AI model is used to identify the second AI model. For example, the identification information of the second AI model can be a number sequence identifier, an letter sequence identifier, or a dataset identifier, where the dataset identifier indicates that the second AI model was trained on a specific dataset. For instance, a device may have multiple datasets, and the device can train multiple AI models based on these datasets. The dataset identifier indicates the dataset used by a particular AI model. As an example, different datasets can be identified using different identifiers.

It should be noted that the identification information of the second AI model can be randomly generated or predefined, and it should be understood that this application does not impose any restrictions on this.

It should be understood that the identification information of the second AI model mentioned above is merely an example, and this application does not impose any limitations on it.

As shown in Table 4, let's take the example of having 3 AI models on the terminal device side.

Table 4

As shown in Table 4, the AI model corresponding to identifier 1 supports 4 resource configurations (resource configuration #1 to resource configuration #4), the AI model corresponding to identifier 2 supports 3 resource configurations (resource configuration #1 to resource configuration #3), and the AI model corresponding to identifier 3 supports 3 resource configurations (resource configuration #1 to resource configuration #3). The description of the resource configurations supported by the AI model corresponding to identifier 1 can be found above and will not be repeated here. The AI model corresponding to identifier 2 supports no restrictions on the offset of adjacent resources in the aperiodic CSI-RS configuration and the number of CSI-RS resource transmissions. The AI model corresponding to identifier 3 supports an offset of m6 for adjacent resources in the aperiodic CSI-RS configuration and a transmission count of K6 for the CSI-RS resource.

Furthermore, the terminal device sends the identification information (e.g., identifier 1) of the first AI model to the network device. Based on the received identifier 1 and the correspondence between the identification information of all AI models on the terminal device side obtained by the protocol predefined and at least one resource configuration, the network device determines at least one resource configuration supported by the first AI model corresponding to identifier 1, and selects one resource configuration (e.g., the first resource configuration) from the at least one resource configuration and sends it to the terminal device.

Method 3:

When the first information includes the identification information of the first AI model, that is, when the terminal device sends the identification information of the first AI model to the network device, optionally, in one possible implementation, the method may further include: the terminal device sending second indication information to the network device, and the network device receiving the second indication information accordingly.

Optionally, in one possible implementation, the second indication information indicates the correspondence between the identification information of the first AI model (e.g., identifier 1) and at least one resource configuration, as shown in Table 3. After receiving the second indication information and the identification information of the first AI model (e.g., identifier 1), the network device further determines, based on the identification information of the first AI model reported by the terminal device (i.e., identifier 1) and the second indication information, at least one resource configuration supported by the first AI model corresponding to identifier 1, and selects one resource configuration (e.g., the first resource configuration) from the at least one resource configuration and sends it to the terminal device.

Alternatively, in another possible implementation, the second indication information may also be used to indicate the correspondence between the identification information of the second AI model and at least one resource configuration supported by the second AI model, wherein the second AI model can be understood as one or more AI models different from the first AI model.

As shown in Table 4, the second indication information indicates the correspondence between the AI model corresponding to identifier 1 and at least one resource configuration, the correspondence between the AI model corresponding to identifier 2 and at least one resource configuration, and the correspondence between the AI model corresponding to identifier 3 and at least one resource configuration. That is, the second indication information simultaneously indicates the correspondence between the identifier information of three AI models and their corresponding at least one resource configuration. Further, after receiving the second indication information and the identifier information of the first AI model (e.g., identifier 1), the network device determines at least one resource configuration supported by the first AI model based on identifier 1 and the second indication information, and selects one resource configuration (e.g., the first resource configuration) from the at least one resource configuration and sends it to the terminal device.

Optionally, in one possible implementation, the terminal device sending the second indication information to the network device as described above can be replaced by the terminal device sending second information to the network device, which includes the second indication information. In other words, the terminal device can report the second indication information to the network device through the second information.

Specifically, the second information may be model-related information, or it may be registration request information, or it may be terminal device capability information. It should be understood that the above are merely illustrative examples, and this application does not impose any limitations on them.

For example, when the second information is model-related information, it can be understood as the second instruction information exchanged by the terminal device during the bilateral model docking phase. For example, when the second information is a registration request message, it can be understood as the second instruction information exchanged by the terminal device during the phase of requesting access to the network device; that is, the terminal device sends a registration request message to the network device requesting registration with the network, and this registration request message includes the second instruction information. For example, when the second information is the terminal device's capability information, it can be understood as the terminal device reporting its own capability information to the network device while simultaneously reporting the second instruction information.

Optionally, in one possible implementation, the first information mentioned above also includes functional information of the first AI model.

Specifically, the terminal device sends first information to the network device, and the network device receives the first information accordingly. This first information also includes functional information of the first AI model. This functional information can be time-domain channel state information (CSI) prediction followed by compression, spatial and frequency-domain CSI compression, or spatial, frequency, and time-domain CSI compression, or time-domain CSI prediction, or beamforming time-domain prediction. It should be understood that the above are merely examples, and this application does not impose any limitations.

It should be understood that the functional information of the first AI model can be understood as the problems that the first AI model can solve, or the tasks that the first AI model can perform.

For example, the first AI model can be used for compression of time-domain channel state information (CSI) prediction. For instance, several historical CSIs or the current CSIs can be input into the first AI model, and the output predicted CSIs for one or more future times can be compressed.

For example, the first AI model can be used for CSI compression in both the spatial and frequency domains, such as jointly compressing the frequency and spatial information corresponding to the first resource configuration based on the first AI model.

For example, the first AI model can be used for CSI compression in the spatial, frequency, and time domains, such as jointly compressing the frequency, spatial, and time domain information corresponding to the first resource configuration based on the first AI model.

For example, the first AI model can be used for CSI prediction of time-domain channel state information. For instance, several historical CSIs or the current CSIs can be input into the first AI model, and the predicted CSIs for one or more future moments can be output.

For example, the first AI model can be used for beam temporal prediction. For instance, it can input first channel information into the first AI model and output second channel information (or predicted channel information). The first channel information is the RSRP in the first beam set, and the second channel information is the RSRP in the second beam set or the identifiers of the optimal K beams, etc. The first and second beam sets can be the same or different. The optimal K beams can be the K beams in the second beam set whose channel quality (e.g., RSRP, SINR) is greater than a certain threshold, where K is a positive integer.

As shown in Table 5, taking the time-domain CSI information compression of the functional information of the first AI model as an example, the terminal device sends the first information to the network device. This first information may include the identification information of the first AI model, the functional information of the first AI model, and the resource configuration supported by the first AI model. The identification information of the first AI model is illustrated below as Identifier 1.

Table 5

As can be seen from Table 5, the functional information of the AI model corresponding to Identifier 1 includes the time-domain CSI prediction and compression function information, and the resource configurations supported by the AI model corresponding to Identifier 1 (resource configuration #1 to resource configuration #4). For the resource configurations supported by the AI model corresponding to Identifier 1, please refer to the previous text, which will not be repeated here.

Optionally, in one possible implementation, the first information mentioned above may further include functional information of the second AI model, wherein the second AI model can be understood as one or more AI models different from the first AI model. This can be understood as the terminal device sending functional information of all AI models supported by the terminal device to the network device. As shown in Table 6, this is illustrated using an example of the terminal device supporting five AI models.

Table 6

As shown in Table 6, the AI model corresponding to identifier 1 provides time-domain CSI prediction and compression functionality; the AI model corresponding to identifier 2 provides spatial and frequency-domain CSI compression functionality; the AI model corresponding to identifier 3 provides spatial, frequency, and time-domain CSI compression functionality; the AI model corresponding to identifier 4 provides time-domain CSI prediction functionality; and the AI model corresponding to identifier 5 provides beamforming time-domain prediction functionality. Specifically, the AI model corresponding to identifier 4 supports an offset of m7 for adjacent CSI-RS resources with an aperiodic configuration and a transmission count of K7 for CSI-RS resources; the AI model corresponding to identifier 5 supports an offset of m8 for adjacent CSI-RS resources with an aperiodic configuration and a transmission count of K8 for CSI-RS resources. It should be noted that the descriptions of the resource configurations supported by the AI models corresponding to identifiers 1 to 3 shown in Table 6 can be found above and will not be repeated here.

It should be understood that Tables 1 to 6 are merely illustrative examples. For instance, the functional information of the AI model corresponding to identifier 1 in Table 6 may be spatial and frequency domain CSI compression functional information, and the functional information of the AI model corresponding to identifier 2 may be time domain CSI compression functional information, etc. This application does not impose any restrictions on this.

Figure 7 is a schematic flowchart of a communication method 700 provided in an embodiment of this application. As shown in Figure 7, the method includes at least the following steps.

Optionally, before step S720, the method includes: S710, the network device sends a second resource configuration to the terminal device, and correspondingly, the terminal device receives the second resource configuration.

S720, the terminal device sends a second channel report and a third indication information to the network device, and the network device receives the second channel report and the third indication information accordingly.

The third indication information instructs the second channel report to be associated with a third AI model. The second channel report is generated by processing channel information measured from the second resource configuration based on the third AI model. Specifically, after receiving the second resource configuration, the terminal device measures the channel information based on the second resource configuration and processes this channel information using the third AI model. Subsequently, the terminal device sends the third indication information to the network device, instructing the second channel report to be associated with the third AI model, so that the network device can use the AI model corresponding to the third AI model to decompress the second channel report to obtain the recovered channel information.

For example, when the third AI model is an AI CSI prediction and compression model, the channel information is processed based on the third AI model, including: performing prediction processing on the channel information based on the third AI model, and then further compressing the channel information obtained from the prediction processing.

For example, when the third AI model is an AI CSI compression model, processing the channel information based on the third AI model includes: compressing the channel information based on the third AI model.

It should be noted that the association of the second channel report with the third AI model can be understood as the second channel report being generated based on the processing of the third AI model. Here, "association" can be replaced with "correspondence", and this application embodiment does not limit this.

Optionally, in one possible implementation, the third indication information can be the identification information of the third AI model. Specifically, after receiving the identification information of the third AI model, the network device knows that the second channel report is generated based on the third AI model, so that the network device can use the AI model corresponding to the third AI model to decompress the second channel report to obtain the recovered channel information.

Optionally, before step S710, the method may further include: the network device determining a second resource configuration. It should be noted that the description of the network device determining the second resource configuration can also refer to the relevant description of the network device determining the first resource configuration described above; for simplicity, it will not be repeated here.

Optionally, in one possible implementation, before step S720, the method may further include: the network device sending fourth indication information to the terminal device, and correspondingly, the terminal device receiving the fourth indication information.

The fourth indication information specifies the configuration information for the second channel report, including its maximum feedback overhead. Specifically, after receiving the fourth indication information, the terminal device learns the maximum feedback overhead of the second channel report, selects the third AI model, processes the channel information obtained from the second resource configuration measurement based on the third AI model, and generates the second channel report.

For example, the configuration information for the second channel report may include a maximum feedback overhead of M bits.

Optionally, in one possible implementation, before step S720, the method may further include: the terminal device determining information for the second channel report based on the second resource configuration. Specifically, the terminal device measures channel information based on the second resource configuration and processes the channel information based on a third AI model to generate the second channel report.

It should be noted that, in the embodiments of this application, the information reported by the second channel includes at least one of the following:

The number of resources corresponding to one or more measurement resources

Measurement results corresponding to one or more measurement resources

Temporal information corresponding to one or more measurement resources

Frequency domain information corresponding to one or more measurement resources

Spatial information corresponding to one or more measurement resources

The number of bits corresponding to each resource information in one or more measurement resources.

It should also be noted that in this embodiment, the terminal device has multiple AI models, and the information in the channel reports generated based on different AI models is different. For example, as shown in Table 7, the example of three AI models on the terminal device side will be used for illustration.

Table 7

As shown in Table 7, the AI model corresponding to identifier 6 has a time-domain prediction and compression function, the AI model corresponding to identifier 7 has a spatial and frequency-domain information compression function, and the AI model corresponding to identifier 8 has a time-domain prediction and spatial, frequency, and time-domain information compression function. Therefore, the channel reports generated by these three AI models based on the channel information obtained from the second resource configuration measurement are also different, as shown in Table 8.

Table 8

As shown in Table 8, the channel report generated based on the AI model corresponding to identifier 6 includes: a predicted CSI for 4 slots, meaning it includes the time-domain information corresponding to the measurement resource for 4 slots; and the total cost M of the channel report, which can include the number of bits corresponding to each resource information in multiple measurement resources, which is M/4. The channel report generated based on the AI model corresponding to identifier 7 includes: a predicted CSI for 1 slot, meaning it includes the time-domain information corresponding to the measurement resource for 1 slot; and the total cost M of the channel report, which can include the number of bits corresponding to each resource information in multiple measurement resources, which is M. The channel report generated based on the AI model corresponding to identifier 8 includes: a predicted CSI for the currently measured slot, meaning it includes the time-domain information corresponding to the measurement resource for the currently measured slot; and the total cost M of the channel report, which can include the number of bits corresponding to each resource information in multiple measurement resources.

It should be understood that Tables 7 and 8 are merely examples and are not intended to limit the scope of this application.

It should be noted that, in one possible scenario, the third AI model can also be the first AI model described above, and correspondingly, the second channel report can also be the first channel report described above, and the second resource configuration can be the first resource configuration described above. In other words, the embodiments shown in Figures 6 and 7 can be used in combination under certain circumstances. As shown in Figure 8, Figure 8 is a schematic flowchart of a communication method 800 provided by an embodiment of this application, which includes at least the following steps.

S810, the network device determines the second resource configuration.

Step S810 is similar to step S610, and will not be described in detail here.

In S820, the network device sends the second resource configuration, and the terminal device receives the second resource configuration accordingly.

Specifically, the terminal device processes the channel information obtained from the second resource configuration measurement based on an AI model (e.g., AI model #3) to generate a second channel report.

S830, the terminal device sends a second channel report and a third indication information to the network device, and the network device receives the second channel report and the third indication information accordingly.

It should be noted that step S830 is similar to step S720, and will not be described in detail here.

Optionally, in one possible implementation, before step S830, the method may further include S840, in which the network device sends fourth indication information to the terminal device, and correspondingly, the terminal device receives the fourth indication information. It should be noted that step S840 can be referenced to the relevant description of the step before step S720 above, in which the network device sends the fourth indication information to the terminal device, and will not be repeated here.

Figure 9 is a schematic flowchart of a communication method 900 provided in an embodiment of this application. As shown in Figure 9, the method includes at least the following steps.

S910, the terminal device sends first information to the network device, and the network device receives the first information accordingly. The first information includes at least one resource configuration supported by the first AI model.

Specifically, the terminal device sends first information to the network device. This first information includes at least one resource configuration supported by the first AI model on the terminal device side. That is, the terminal device reports at least one resource configuration supported by the first AI model to the network device. It should be noted that a detailed description of the at least one resource configuration supported by the first AI model reported by the terminal device can be found in the relevant description in Table 1, and will not be repeated here.

Optionally, in one possible implementation, the first information may also include the identification information of the first AI model. That is, in this case, the first information may include the identification information of the first AI model and at least one resource configuration supported by the first AI model. For example, the specific content of the first information can be found in Table 3, and will not be elaborated here.

Optionally, in one possible implementation, the first information may also include functional information of the first AI model. That is, in this case, the first information may include the identification information of the first AI model, the functional information of the first AI model, and at least one resource configuration supported by the first AI model. For example, the specific content of the first information can be found in Table 5, and will not be elaborated here.

Optionally, in one possible implementation, the first information may further include at least one resource configuration supported by the second AI model, wherein the second AI model can be understood as one or more AI models different from the first AI model.

Specifically, the terminal device sends first information to the network device. This first information includes at least one resource configuration supported by a first AI model on the terminal device side, and at least one resource configuration supported by other AI models (second AI models) on the terminal device side. That is, the terminal device reports to the network device at least one resource configuration supported by the first AI model and at least one resource configuration supported by other AI models. Optionally, in some cases, the terminal device reports to the network device at least one resource configuration supported by all AI models on the terminal device side.

It should be noted that for a detailed description of at least one resource configuration supported by the first AI model reported by the terminal device and at least one resource configuration supported by other AI models, please refer to the relevant descriptions in the preceding text and Table 2, which will not be repeated here.

Optionally, in one possible implementation, the first information may further include the identification information of the first AI model and the identification information of the second AI model. That is, in this case, the first information may include the identification information of the first AI model, at least one resource configuration supported by the first AI model, the identification information of the second AI model, and at least one resource configuration supported by the second AI model. For example, the specific content of the first information can be found in Table 4, and will not be elaborated here.

Optionally, in one possible implementation, the first information may further include functional information of the first AI model and functional information of the second AI model. That is, in this case, the first information may include the identification information of the first AI model, the functional information of the first AI model, at least one resource configuration supported by the first AI model, the identification information of the second AI model, the functional information of the second AI model, and at least one resource configuration supported by the second AI model. For example, the specific content of the first information can be found in Table 6, and will not be elaborated here.

In S920, the network device sends a first resource configuration to the terminal device, and the terminal device receives the first resource configuration accordingly.

Specifically, after receiving the first information, the network device determines at least one resource configuration supported by the AI model on the terminal device side, and selects one resource configuration from the at least one resource configuration supported by the AI model on the terminal device side and sends it to the terminal device. For example, the network device selects a first resource configuration to send to the terminal device, wherein the first resource configuration is one of the at least one resource configuration supported by the first AI model. In other words, the terminal device selects a first resource configuration from the at least one resource configuration supported by the first AI model and sends it to the terminal device.

S930, the network device sends a reference signal CSI-RS to the terminal device, and the terminal device receives the reference signal CSI-RS accordingly.

S940, the terminal device performs measurements based on the received first resource configuration to obtain channel information.

S950: The terminal device processes the channel information based on the first AI model to obtain the first channel report.

The first AI model can be either an AI CSI prediction post-compressed model or an AI CSI compressed model.

For example, when the first AI model is an AI CSI compression model, the terminal device processes the channel information based on the first AI model, including: the terminal device compresses the channel information based on the first AI model. That is, in this example, the first channel report is generated based on the compression processing of the channel information using the first AI model.

For example, when the first AI model is an AI CSI prediction and compression model, the terminal device processes the channel information based on the first AI model, including: the terminal device performs prediction processing on the channel information based on the first AI model, and then further compresses the channel information obtained from the prediction processing. That is, in this example, the first channel report is generated based on the prediction processing of the channel information using the first AI model, followed by further compression processing of the channel information obtained from the prediction processing.

S960, the terminal device sends a first channel report to the network device, and the network device receives the first channel report accordingly.

S970, the network device decompresses the first channel report based on AI model*1 to obtain the recovered channel information.

Among them, AI model *1 is matched with the first AI model, or in other words, AI model *1 and the first AI model are in one-to-one correspondence.

According to the above technical solution, the terminal device reports at least one resource configuration supported by the AI model on the terminal device side to the network device, which can guarantee or improve the performance of the AI model when the network device makes different resource configurations to the terminal device, and further guarantee the feedback performance of CSI.

Figure 10 is a schematic flowchart of a communication method 1000 provided in an embodiment of this application. As shown in Figure 10, the method includes at least the following steps. It should be noted in advance that steps S1010 to S1040 are similar to steps S910 to S940 described above. For simplicity, only the differences from the embodiment shown in Figure 9 above will be described below, and will not be repeated here.

S1050, the terminal device predicts the channel information based on the first AI model to obtain the predicted channel information.

The first AI model can be either an AI CSI prediction model or an AI beam temporal prediction model.

According to the above technical solution, the terminal device reports at least one resource configuration supported by the AI model on the terminal device side to the network device, which can guarantee or improve the predictive performance of the AI model when the network device makes different resource configurations to the terminal device.

Figure 11 is a schematic flowchart of a communication method 1100 provided in an embodiment of this application. As shown in Figure 11, the method includes at least the following steps.

S1110, the network device sends the second resource configuration to the terminal device, and the terminal device receives the second resource configuration accordingly.

S1120, the network device sends the configuration information of the second channel report to the terminal device, and the terminal device receives the configuration information of the second channel report accordingly.

Specifically, the configuration information of the second channel report includes the maximum feedback overhead of the second channel report. After receiving the configuration information of the second channel report, the terminal device knows the maximum feedback overhead of the second channel report.

S1130, the network device sends a reference signal CSI-RS to the terminal device, and the terminal device receives the reference signal CSI-RS accordingly.

S1140, the terminal device performs measurements based on the second resource configuration to obtain channel information.

S1150, the terminal device processes the channel information based on the third AI model to obtain the second channel report.

Specifically, when the terminal device learns the maximum feedback cost of the second channel report, it selects a third AI model based on the maximum feedback cost of the second channel report, and processes the channel information obtained from the second resource configuration measurement based on the third AI model to generate the second channel report. The third AI model can be an AI CSI prediction-compressed model or an AI CSI compressed model.

For example, when the third AI model is an AI CSI compression model, the terminal device compresses the channel information using the third AI model and then generates a second channel report.

For example, when the first AI model is an AI CSI prediction and compression model, the terminal device performs prediction processing on the channel information using the first AI model, and then further compresses the channel information obtained from the prediction processing to generate a second channel report.

S1160, the terminal device sends a second channel report and a third indication information to the network device, and the network device receives the second channel report and the third indication information accordingly.

Specifically, the third instruction information instructs the second channel report to be associated with the third AI model. Specifically, the terminal device sends the third instruction information to the network device, instructing the second channel report to be associated with the third AI model. After receiving the third instruction information and the second channel report, the network device knows that the second channel report is generated based on the third AI model. This allows the network device to use the AI model corresponding to the third AI model to decompress the second channel report and obtain the recovered channel information.

S1170, the network device decompresses the second channel report based on AI model*2 to obtain the recovered channel information.

Among them, AI Model *2 is matched with the third AI model, or in other words, AI Model *2 corresponds one-to-one with the third AI model.

It should be noted that in this embodiment, the terminal device may have multiple AI models, and the information in the second channel report generated based on different AI models will be different. For example, specific descriptions can be found in Tables 7 and 8 above, which will not be repeated here.

According to the above technical solution, the terminal device matches the configuration information of the channel report sent by the network device in real time, selects the most suitable AI model, which can guarantee or improve the performance of the AI model when the network device makes different resource configurations to the terminal device, and further guarantee the performance of CSI feedback.

Figures 6 to 11 illustrate the scenario where the first device is the terminal device and the second device is the network device. The following description, in conjunction with Figures 12 to 14, illustrates scenarios where the first and second devices serve as AI entities for the terminal device and network device, respectively. The AI entity on the terminal device side can be the terminal device itself or an AI entity serving the terminal device, such as a server, like an OTT server or a cloud server. The AI entity on the network device side can be the network device itself or an AI entity serving the network device, such as RAN, RIC, OAM, or a server, like a cloud server. The AI entity on the network device can also be replaced by intelligent network elements, etc., without limitation. It should be noted that intelligent network elements can also be understood as network devices with AI capabilities, applicable in both O-RAN and non-O-RAN architectures, without limitation.

Figure 12 is a schematic flowchart of another communication method 1200 according to an embodiment of this application. As shown in Figure 12, the first device is an OTT, the second device is a near real-time RIC, and the method includes:

S1201, the OTT sends the first information to the terminal device, and the terminal device receives the first information accordingly.

S1202, the terminal device sends the first information to the network device, and the network device receives the first information accordingly.

S1203, the network device sends the first information to the near real-time RIC, and correspondingly, the near real-time RIC receives the first information.

S1204, the network device sends the first resource configuration to the terminal device, and the terminal device receives the first resource configuration accordingly.

S1205, the network device sends a reference signal CSI-RS to the terminal device, and the terminal device receives the reference signal CSI-RS accordingly.

S1206, the terminal device performs measurements based on the received first resource configuration to obtain channel information.

For a description of steps S1204 to S1206, please refer to steps S920 to S940, which will not be repeated here.

S1207, the terminal device sends channel information to the OTT, and the OTT receives the channel information accordingly.

S1208, OTT processes the channel information based on the first AI model to obtain the first channel report.

For a description of step S1208, please refer to the description of step S950 above, which will not be repeated here.

S1209, the OTT sends a first channel report to the terminal device, and the terminal device receives the first channel report accordingly.

S1210, the terminal device sends a first channel report to the network device, and the network device receives the first channel report accordingly.

S1211, the network device sends the first channel report to the near real-time RIC, and correspondingly, the near real-time RIC receives the first channel report.

S1212, the near real-time RIC decompresses the first channel report based on AI model*1 to obtain the recovered channel information.

Among them, AI model *1 is matched with the first AI model, or in other words, AI model *1 and the first AI model are in one-to-one correspondence.

Figure 13 is a schematic flowchart of another communication method 1300 according to an embodiment of this application. As shown in Figure 13, the first device is an OTT and the second device is a near real-time RIC. The method includes the following steps. It should be noted in advance that steps S1301 to S1307 are similar to steps S1201 to S1207 described above. For simplicity, only the differences from the embodiment shown in Figure 12 above will be described below, and will not be repeated here.

S1308, OTT predicts channel information based on the first AI model to obtain the predicted channel information.

The first AI model can be either an AI CSI prediction model or an AI beam temporal prediction model.

Figure 14 is a schematic flowchart of another communication method 1400 according to an embodiment of this application. As shown in Figure 14, the first device is an OTT, the second device is a near real-time RIC, and the method includes:

S1401, the network device sends the second resource configuration to the terminal device, and the terminal device receives the second resource configuration accordingly.

S1402, the network device sends the configuration information of the second channel report to the terminal device, and the terminal device receives the configuration information of the second channel report accordingly.

For a description of S1402, please refer to the description of step S1120 above, which will not be repeated here.

S1403, the network device sends a reference signal CSI-RS to the terminal device, and the terminal device receives the reference signal CSI-RS accordingly.

S1404, the terminal device performs measurements based on the second resource configuration to obtain channel information.

S1405, the terminal device sends channel information to the OTT, and the OTT receives the channel information accordingly.

S1406, OTT processes the channel information based on the third AI model to obtain the second channel report.

For a description of step S1406, please refer to the description of step S1150 above, which will not be repeated here.

S1407, the OTT sends a second channel report and a third indication information to the terminal device, and the terminal device receives the second channel report and the third indication information accordingly.

S1408, the terminal device sends a second channel report and a third indication information to the network device, and the network device receives the second channel report and the third indication information accordingly.

For a description of steps S1407 and S1408, please refer to the description of step S1160 above, which will not be repeated here.

S1409, the network device sends a second channel report and a third indication information to the near real-time RIC, and the near real-time RIC receives the second channel report and the third indication information accordingly.

S1410, the near real-time RIC decompresses the second channel report based on AI model*2 to obtain the recovered channel information.

Among them, AI Model *2 is matched with the third AI Model, or in other words, AI Model *2 corresponds one-to-one with the third AI Model.

Finally, the device embodiments of this application will be described.

To achieve the functions of the methods provided in this application, both the first device and the second device may include hardware structures and/or software modules, implementing the aforementioned functions in the form of hardware structures, software modules, or a combination of hardware structures and software modules. Whether a particular function is executed in the form of hardware structures, software modules, or a combination of hardware structures and software modules depends on the specific application and design constraints of the technical solution.

Figure 15 is a schematic block diagram of a communication device according to an embodiment of this application. The communication device includes a processing circuit 1510 and a transceiver circuit 1520, which can be interconnected or coupled to each other, for example, through a bus 1530. The communication device can be a first device or a second device.

Optionally, the communication device may also include a memory 1540. The memory 1540 includes, but is not limited to, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), or compact disc read-only memory (CD-ROM), which is used for related instructions and data.

The processing circuit 1510 may be all or part of the processing or control circuitry of one or more processors, or it may be one or more processors. The processor may be a central processing unit (CPU). If the processing circuit 1510 is a CPU, the CPU may be a single-core CPU or a multi-core CPU.

The processing circuit 1510 may be a signal processor, a chip, or other integrated circuit that can implement the method of this application, or a portion of the aforementioned processor, chip, or integrated circuit used for processing functions.

The transceiver circuit 1520 can also be a transceiver, or an input/output interface. An input/output interface is used for the input or output of signals or data and can also be called an input/output circuit.

When the communication device is the first device, the transceiver circuit 1520 is configured to perform the following operations: receive first indication information, which indicates a first resource configuration, which is one of at least one resource configuration supported by a first artificial intelligence (AI) model.

Optionally, in some examples, the processing circuit 1510 is used to determine a first channel report, which is determined based on the channel information obtained by the first AI model from the measurement of the first resource configuration.

When the communication device is a second device, the processing circuit 1510 is configured to perform the following operations: determine a first resource configuration, which is one of at least one resource configuration supported by a first AI model, the first AI model being configured to process channel information measured based on the first resource configuration.

The transceiver circuit 1520 is configured to perform the following operations: send a first indication message that indicates the first resource configuration.

The above description is for illustrative purposes only. When the communication device is the first device or the second device, it will be responsible for executing the methods or steps related to the first device or the second device in the foregoing method embodiments.

When the communication device is the first device or the second device, the transceiver circuit 1520 can be a transceiver or an interface circuit.

When the communication device is a chip used in the first device or the second device, the transceiver circuit 1520 can be an input/output circuit.

For details, please refer to the content shown in the above method embodiments.

The implementation of each operation in Figure 15 can also be described in the corresponding description of the method embodiments shown in Figures 6 to 14.

Figure 16 is another schematic block diagram of a communication device according to an embodiment of this application. This communication device can be a first device or a second device, used to implement the methods involved in the above embodiments.

The communication device includes a transceiver unit 1610 and a processing unit 1620. The transceiver unit 1610 may include a sending unit and a receiving unit. The sending unit is used to perform the sending action of the communication device, and the receiving unit is used to perform the receiving action of the communication device. For ease of description, the sending unit and the receiving unit are combined into one transceiver unit in this embodiment. This will be explained uniformly here and will not be repeated later.

When the communication device is the first device, exemplarily, the transceiver unit 1610 is used to receive first instruction information, etc.; the processing unit 1620 is used to execute the processing, control, and other steps involved in the first device. For example, the processing unit 1620 is used to determine a first channel report, etc.

When the communication device is a second device, exemplarily, the transceiver unit 1610 is used to send first instruction information, etc.; the processing unit 1620 is used to execute the contents of the second device involving processing, control, etc. For example, the processing unit 1620 is used to determine a first resource configuration, etc.

When the communication device is the first device or the second device, it will be responsible for executing one or more of the methods or steps related to the first device or the second device in the foregoing method embodiments.

Optionally, the communication device further includes a storage unit 1630 for storing programs or code for performing the aforementioned methods.

The transceiver unit in Figure 16 can correspond to the transceiver circuit in Figure 15, and the processing unit in Figure 16 can correspond to the processing circuit in Figure 15.

The apparatus embodiments shown in Figures 15 and 16 are used to implement the contents described in Figures 6 to 14. The specific execution steps and methods of the apparatus shown in Figures 15 and 16 can be found in the foregoing method embodiments.

This application also provides a chip, including a processor, for calling and executing instructions stored in a memory, causing a communication device on which the chip is installed to perform the methods described in the examples above. The memory may be integrated within the chip or located externally.

This application also provides another chip, including: an input interface, an output interface, and a processing circuit. The input interface, the output interface, and the processor are connected via an internal connection path. The processing circuit is used to execute code in a memory. When the code is executed, the processing circuit is used to execute the methods in the examples described above. Optionally, the chip also includes a memory for storing computer programs or code. The input interface and the output interface can be independent of each other, or they can be integrated into a single input/output interface.

The processing circuitry can be all or part of the processing circuitry in one or more processors, or one or more processors.

This application also provides a processor for coupling with a memory for performing the methods and functions involving the first or second apparatus in any of the above embodiments.

In another embodiment of this application, a computer program product containing instructions is provided, which, when run on a computer, enables the implementation of the methods described in the foregoing embodiments.

This application also provides a computer program that, when run on a computer, enables the implementation of the methods described in the foregoing embodiments.

In another embodiment of this application, a computer-readable storage medium is provided, which stores a computer program that, when executed by a computer, implements the methods described in the foregoing embodiments.

It should be understood that in the embodiments of this application, the processor can be a central processing unit (CPU), but it can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor.

It should also be understood that the memory in the embodiments of this application can be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. The non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced synchronous DRAM (ESDRAM), synchronous linked DRAM (SLDRAM), and direct rambus RAM (DR RAM). It should be noted that the memory used in the systems and methods described herein is intended to include, but is not limited to, these and any other suitable types of memory.

The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. A computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or 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 website, computer, server, or data center via wired or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. A semiconductor medium can be a solid-state drive.

It should be understood that in the various embodiments of this application, the order of the above-mentioned processes does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

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. Those skilled in the art will clearly 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. In the several embodiments provided in this application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative; for example, the division of units is merely 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 mutual coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection of devices or units may be electrical, mechanical, or other forms.

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. Furthermore, 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. If the above 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 part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions 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, random access memory, magnetic disks, or optical disks.

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 (37)

A communication method, characterized in that it includes: Receive a first instruction message, the first instruction message indicating a first resource configuration, the first resource configuration being one of at least one resource configuration supported by a first artificial intelligence (AI) model. The first AI model is used to process the channel information obtained based on the first resource configuration measurement. The method according to claim 1, wherein the resource configuration includes: Configuration type, offset of adjacent resources, or number of transmissions. The configuration type includes at least one of the following: Periodic configuration, semi-static configuration, or non-periodic configuration. The method according to claim 1 or 2, characterized in that, the first AI model is used to process the channel information obtained based on the first resource configuration measurement, and further includes: A first channel report is determined, which is based on the channel information obtained by the first AI model from the first resource configuration measurement. The method according to any one of claims 1 to 3, characterized in that the method further comprises: Send a first message, which is associated with the at least one resource configuration. The method according to claim 4, wherein the first information includes at least one resource configuration. The method according to claim 4, wherein the first information includes the identification information of the first AI model. The method according to claim 6, characterized in that the method further comprises: Send a second instruction message, which indicates the correspondence between the identification information of the first AI model and the at least one resource configuration. The method according to claim 7, characterized in that, The second instruction information also indicates the correspondence between the identification information of the second AI model and at least one resource configuration supported by the second AI model, wherein the second AI model is different from the first AI model. The method according to claim 7 or 8, wherein sending the second indication information includes: Send a second message, the second message including the second indication message, the second message including any one of the following: Model-related information, registration request information, and terminal device capability information. The method according to any one of claims 4 to 9, characterized in that the first information includes functional information of the first AI model, and the functional information includes any one of the following: Time-domain channel state information, CSI prediction post-compression function information, Spatial and frequency domain CSI compression function information, Spatial, frequency, and time domain CSI compression function information Time-domain CSI prediction function information Beam temporal prediction function information. The method according to any one of claims 1 to 10, characterized in that it comprises: Send a second channel report and a third indication information, wherein the third indication information indicates that the second channel report is associated with a third AI model, and the second channel report is generated by processing the channel information obtained from the second resource configuration measurement based on the third AI model. The method according to claim 11, characterized in that the method further comprises: Receive a fourth indication message, which indicates the configuration information of the second channel report, including the maximum feedback overhead of the second channel report. The method according to claim 11 or 12, characterized in that the method further comprises: The information in the second channel report is determined based on the second resource configuration. The method according to any one of claims 11 to 13, characterized in that the information reported by the second channel includes at least one of the following: The number of resources corresponding to one or more measurement resources Measurement results corresponding to one or more measurement resources Temporal information corresponding to one or more measurement resources Frequency domain information corresponding to one or more measurement resources Spatial information corresponding to one or more measurement resources The number of bits corresponding to each resource information in one or more measurement resources. A communication method, characterized in that it includes: A first resource configuration is determined, wherein the first resource configuration is one of at least one resource configuration supported by a first AI model, and the first AI model is used to process channel information measured based on the first resource configuration; Send a first instruction message, which indicates the first resource configuration. The method according to claim 15, wherein the resource configuration includes: Configuration type, offset of adjacent resources, or number of transmissions. The configuration type includes at least one of the following: Periodic configuration, semi-static configuration, or non-periodic configuration. The method according to claim 15 or 16, characterized in that the method further comprises: Obtain first information, which is associated with at least one resource configuration. The method according to claim 17, wherein the first information includes the at least one resource configuration. The method according to claim 17, wherein the first information includes the identification information of the first AI model. The method according to claim 19, characterized in that the method further comprises: Obtain second indication information, which indicates the correspondence between the identification information of the first AI model and the at least one resource configuration. The method according to claim 20, characterized in that, The second instruction information also indicates the correspondence between the identification information of the second AI model and at least one resource configuration supported by the second AI model, wherein the second AI model is different from the first AI model. The method according to claim 20 or 21, wherein receiving the second indication information includes: Obtain second information, the second information including the second indication information, the second information including any one of the following: Model-related information, registration request information, and terminal device capability information. The method according to any one of claims 17 to 22, characterized in that the first information includes functional information of the first AI model, and the functional information includes any one of the following: Information on time-domain CSI prediction followed by compression function Spatial and frequency domain CSI compression function information, Spatial, frequency, and time domain CSI compression function information Time-domain CSI prediction function information Beam temporal prediction function information. The method according to any one of claims 15 to 23, characterized in that it comprises: Obtain a second channel report and a third indication information. The third indication information indicates that the second channel report is associated with a third AI model. The second channel report is generated by processing the channel information obtained from the second resource configuration measurement based on the third AI model. The method according to claim 24, characterized in that the method further comprises: Send a fourth indication message, which indicates the configuration information of the second channel report, including the maximum feedback overhead of the second channel report. The method according to claim 24 or 25, wherein the information in the second channel report includes at least one of the following: The number of resources corresponding to one or more measurement resources Measurement results corresponding to one or more measurement resources Temporal information corresponding to one or more measurement resources Frequency domain information corresponding to one or more measurement resources Spatial information corresponding to one or more measurement resources The number of bits corresponding to each resource information in one or more measurement resources. A communication method, characterized in that it includes: Send a second channel report and a third indication information, wherein the third indication information indicates that the second channel report is associated with a third AI model, and the second channel report is generated by processing the channel information obtained from the second resource configuration measurement based on the third AI model. The method according to claim 27, characterized in that the method further comprises: Receive a fourth indication message, which indicates the configuration information of the second channel report, including the maximum feedback overhead of the second channel report. The method according to claim 27 or 28, characterized in that the method further comprises: The information in the second channel report is determined based on the second resource configuration. The method according to any one of claims 27 to 29, characterized in that the information reported by the second channel includes at least one of the following: The number of resources corresponding to one or more measurement resources Measurement results corresponding to one or more measurement resources Temporal information corresponding to one or more measurement resources Frequency domain information corresponding to one or more measurement resources Spatial information corresponding to one or more measurement resources The number of bits corresponding to each resource information in one or more measurement resources. A communication method, characterized in that it includes: Obtain a second channel report and a third indication information. The third indication information indicates that the second channel report is associated with a third AI model. The second channel report is generated by processing the channel information obtained from the second resource configuration measurement based on the third AI model. The method according to claim 31, characterized in that the method further comprises: Send a fourth indication message, which indicates the configuration information of the second channel report, including the maximum feedback overhead of the second channel report. The method according to claim 31 or 32, wherein the information reported by the second channel includes at least one of the following: The number of resources corresponding to one or more measurement resources Measurement results corresponding to one or more measurement resources Temporal information corresponding to one or more measurement resources Frequency domain information corresponding to one or more measurement resources Spatial information corresponding to one or more measurement resources The number of bits corresponding to each resource information in one or more measurement resources. A communication device, characterized in that it includes a module or unit for implementing the method of any one of claims 1 to 14, or a module or unit for implementing the method of any one of claims 15 to 26, or a module or unit for implementing the method of any one of claims 27 to 30, or a module or unit for implementing the method of any one of claims 31 to 33. The communication device according to claim 34 is characterized in that, When the communication device includes a module or unit for implementing the method of any one of claims 1 to 14, or a module or unit for implementing the method of any one of claims 27 to 30, the communication device is a terminal device, or... The communication device is a chip used in terminal equipment, or... The communication device is a server on the terminal device side, or; When the communication device includes a module or unit for implementing the method of any one of claims 15 to 26, or a module or unit for implementing the method of any one of claims 31 to 33, the communication device is a network device, or... The communication device is a chip used in network equipment, or... The communication device is a server on the network equipment side. A communication system, characterized in that it includes means for performing the method as described in any one of claims 1 to 14 and means for implementing the method as described in any one of claims 15 to 26, or means for implementing the method as described in any one of claims 27 to 30 and means for implementing the method as described in any one of claims 31 to 33. A readable storage medium, characterized in that it is used to store instructions, when the instructions are executed, the method as described in any one of claims 1 to 33 is performed.