AI-based method, apparatus, medium, product and chip for determining csi processing capability

Abstract

The present disclosure discloses an AI-based CSI processing capability determination method, device, medium, product and chip, and belongs to the field of communication. The method comprises: receiving a reported static capability of a terminal based on artificial intelligence (AI); and determining an AI-based channel state indication (CSI) processing capability of the terminal based on the static capability. The method is used to support the network device to judge the capability of the terminal, so as to configure the AI-based CSI compression matched with the capability of the terminal.

Description

Technical Field

[0001] This disclosure relates to the field of communications, and in particular to an AI-based method, apparatus, medium, product, and chip for determining CSI processing capabilities. Background Technology

[0002] In 5G New Radio systems, Channel Status Indicator (CSI) can be used to obtain information such as the number of information streams a channel can carry, channel quality or signal-to-noise ratio, and channel matrix. Therefore, the acquisition and feedback of CSI is crucial.

[0003] In response to CSI feedback, the 3rd Generation Partner Project (3GPP) standardized Type I and Type II codebooks; furthermore, it introduced Artificial Intelligence (AI) technology to achieve CSI feedback with arbitrary feedback bits and arbitrary precision requirements through AI networks.

[0004] In the AI-based CSI processing, there is an important scenario: CSI compression. The terminal can input the measured full-channel information or feature vectors into the AI ​​model for compression, that is, to achieve AI-based CSI compression. Summary of the Invention

[0005] This disclosure provides an AI-based method, apparatus, medium, product, and chip for determining CSI processing capability. The technical solution is as follows:

[0006] According to one aspect of the present disclosure, an AI-based method for determining CSI processing capability is provided, the method being performed by a network device, the method comprising:

[0007] The static capabilities of the receiving terminal based on AI;

[0008] The terminal's AI-based CSI processing capability is determined based on the static capabilities.

[0009] According to another aspect of the present disclosure, an AI-based method for determining CSI processing capability is provided, the method being executed by a terminal, the method comprising:

[0010] The terminal's AI-based static capabilities are reported to the network device, and these AI-based static capabilities are used to determine the AI-based CSI processing capabilities.

[0011] According to another aspect of the present disclosure, an AI-based CSI processing capability determination apparatus is provided, the apparatus comprising:

[0012] The first receiving module is configured to receive the reported AI-based static capabilities of the terminal.

[0013] The first processing module is configured to determine the terminal's AI-based CSI processing capability based on the static capabilities.

[0014] According to another aspect of the present disclosure, an AI-based CSI processing capability determination apparatus is provided, the apparatus comprising:

[0015] The second sending module is configured to report the terminal's AI-based static capabilities to the network device, the AI-based static capabilities being used to determine the AI-based CSI processing capabilities.

[0016] According to another aspect of the present disclosure, a network device is provided, the network device comprising:

[0017] processor;

[0018] A transceiver connected to the processor;

[0019] The processor is configured to load and execute executable instructions to implement the AI-based CSI processing capability determination method as described in the above aspects.

[0020] According to another aspect of the embodiments of this disclosure, a terminal is provided, the terminal comprising:

[0021] processor;

[0022] A transceiver connected to the processor;

[0023] The processor is configured to load and execute executable instructions to implement the AI-based CSI processing capability determination method as described in the above aspects.

[0024] According to another aspect of the embodiments of this disclosure, a computer-readable storage medium is provided, wherein at least one instruction, at least one program, code set, or instruction set is stored therein, wherein the at least one instruction, the at least one program, the code set, or the instruction set is loaded and executed by a processor to implement the AI-based CSI processing capability determination method as described in the foregoing aspects.

[0025] According to another aspect of the embodiments of this disclosure, a computer program product (or computer program) is provided, the computer program product (or computer program) including computer instructions stored in a computer-readable storage medium; a processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, causing the computer device to perform the AI-based CSI processing capability determination method as described in the foregoing aspects.

[0026] According to another aspect of the embodiments of this disclosure, a chip is provided, the chip including programmable logic circuitry and / or program instructions, which, when the chip is running, are used to implement the AI-based CSI processing capability determination method as described in the foregoing aspects.

[0027] The technical solutions provided in this disclosure may have the following beneficial effects:

[0028] In the above-mentioned AI-based CSI processing capability determination method, the network device can determine the terminal's AI-based CSI processing capability based on the terminal's own AI-based static capabilities reported by the terminal. This method provides a relevant capability reporting mechanism for determining the terminal's CSI processing capability, which supports the network device in judging the terminal's capabilities, thereby configuring AI-based CSI compression that matches the terminal's capabilities.

[0029] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0030] To more clearly illustrate the technical solutions in the embodiments of this disclosure, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this disclosure. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0031] Figure 1 This is a block diagram of a communication system according to an exemplary embodiment;

[0032] Figure 2 This is a flowchart illustrating an AI-based CSI processing capability determination method according to an exemplary embodiment;

[0033] Figure 3 This is a flowchart illustrating an AI-based CSI processing capability determination method according to another exemplary embodiment;

[0034] Figure 4 This is a flowchart illustrating an AI-based CSI processing capability determination method according to another exemplary embodiment;

[0035] Figure 5 This is a flowchart illustrating a method for switching CSI processing modes according to an exemplary embodiment;

[0036] Figure 6 This is a flowchart illustrating a method for switching CSI processing modes according to another exemplary embodiment;

[0037] Figure 7 This is a flowchart illustrating an AI-based CSI processing capability determination method according to another exemplary embodiment;

[0038] Figure 8 This is a flowchart illustrating a method for switching CSI processing modes according to another exemplary embodiment;

[0039] Figure 9 This is a flowchart illustrating a method for switching CSI processing modes according to another exemplary embodiment;

[0040] Figure 10 This is a block diagram illustrating an AI-based CSI processing capability determination apparatus according to an exemplary embodiment;

[0041] Figure 11 This is a block diagram of an AI-based CSI processing capability determination apparatus according to another exemplary embodiment;

[0042] Figure 12 This is a schematic diagram of the structure of a terminal according to an exemplary embodiment;

[0043] Figure 13 This is a schematic diagram of the structure of a network device according to an exemplary embodiment. Detailed Implementation

[0044] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this disclosure. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this disclosure as detailed in the appended claims.

[0045] Figure 1 A block diagram of a communication system provided in an exemplary embodiment of the present disclosure is shown. The communication system may include an access network 12 and a user terminal 14.

[0046] Access network 12 includes a plurality of network devices 120. Network devices (also known as access network devices) 120 may be base stations, which are devices deployed in the access network to provide wireless communication functions to user terminals (hereinafter referred to as "terminals") 14. Base stations may include various forms of macro base stations, micro base stations, relay stations, access points, etc. In systems employing different wireless access technologies, the names of devices with base station functions may differ; for example, in Long Term Evolution (LTE) systems, they are called eNodeB or eNB; in 5G NR (New Radio) systems, they are called gNodeB or gNB. As communication technologies evolve, the description of "base station" may change. For the convenience of the description in the embodiments of this disclosure, the devices that provide wireless communication functions to user terminals 14 are collectively referred to as network devices.

[0047] User terminal 14 may include various handheld devices, vehicle-mounted devices, wearable devices, computing devices, or other processing devices connected to a wireless modem, as well as various forms of user equipment, mobile stations (MS), terminal devices, etc. For ease of description, the devices mentioned above are collectively referred to as user terminals. Network device 120 and user terminal 14 communicate with each other through some air interface technology, such as a Uu interface.

[0048] For example, there are two communication scenarios between network device 120 and user terminal 14: uplink communication scenario and downlink communication scenario. Uplink communication refers to sending signals to network device 120; downlink communication refers to sending signals to user terminal 14.

[0049] The technical solutions of this disclosure can be applied to various communication systems, such as: Global System of Mobile Communication (GSM) system, Code Division Multiple Access (CDMA) system, Wideband Code Division Multiple Access (WCDMA) system, General Packet Radio Service (GPRS), Long Term Evolution (LTE) system, LTE Frequency Division Duplex (FDD) system, LTE Time Division Duplex (TDD) system, Advanced Long Term Evolution (LTE-A) system, New Radio (NR) system, evolution of NR system, LTE-based access to Unlicensed spectrum (LTE-U) system, NR-U system, Universal Mobile Telecommunication System (UMTS), Worldwide Interoperability for Microwave Access (WiMAX) communication system, and Wireless Local Area Network (WLAN). Networks (WLAN), Wireless Fidelity (WiFi), next-generation communication systems, or other communication systems.

[0050] Traditional communication systems typically support a limited number of connections and are easy to implement. However, with the development of communication technology, mobile communication systems will not only support traditional communication but also, for example, device-to-device (D2D) communication, machine-to-machine (M2M) communication, machine-type communication (MTC), vehicle-to-vehicle (V2V) communication, and vehicle-to-everything (V2X) systems. The embodiments disclosed herein can also be applied to these communication systems.

[0051] Figure 2A flowchart of an AI-based CSI processing capability determination method provided in an exemplary embodiment of this disclosure is shown, which is applied to... Figure 1 In the network device of the communication system shown, the method includes:

[0052] Step 210: Receive the reported AI-based static capabilities of the terminal.

[0053] Among them, the static capabilities based on AI are used to indicate the terminal's AI-based computing capabilities.

[0054] Optionally, the terminal's AI-based static capabilities include at least one of the following:

[0055] • Hardware information with AI processing capabilities;

[0056] The hardware information indicating AI processing capabilities is used to indicate the AI ​​processing capabilities of the terminal. For example, the hardware information indicating AI processing capabilities includes hardware information used by the terminal to process CSI compression; for instance, the hardware information used by the terminal to process CSI compression includes at least one of the following: information about the AI ​​processing chip and information about the graphics processing unit (GPU).

[0057] Hardware information with AI processing capabilities can also be used to indicate the level of AI processing capabilities possessed by the terminal. For example, the hardware information used by the terminal to process CSI compression includes the hardware capability level. For instance, the hardware capability level can be defined based on the hardware's performance parameters, that is, the hardware level can be defined based on the hardware capabilities; the terminal determines and reports the hardware capability level based on the hardware's performance parameters.

[0058] Alternatively, the hardware information used by the terminal to process CSI compression includes hardware performance parameters, such as the performance parameters of the AI ​​processing chip and the graphics processing unit (GPU). Then, the network device can directly use these hardware performance parameters to determine the terminal's AI-based CSI processing capability; or, the network device can determine the hardware capability level based on the hardware performance parameters, and then determine the terminal's AI-based CSI processing capability based on the hardware capability level. That is, the information about the AI ​​processing chip can include its performance parameters, and the information about the graphics processing unit can include its performance parameters.

[0059] • Support status of the AI ​​processing platform;

[0060] The AI ​​processing platform support information indicates which AI processing platforms the terminal supports. Specifically, the AI ​​processing platform provides the terminal with AI-based CSI processing capabilities; that is, if the terminal supports an AI processing platform, it can perform AI-based CSI processing through that platform.

[0061] For example, AI processing platforms include TensorFlow, an end-to-end open-source machine learning platform.

[0062] For example, the communication protocol predefines at least one AI processing platform / framework; if the support status of the AI ​​processing platform indicates that the terminal supports any one of the at least one AI processing platform / framework, the network device can determine that the terminal supports AI-based CSI processing, that is, the terminal has AI-based CSI processing capability.

[0063] For example, the support status of an AI processing platform can be reported by the terminal through a bitmap.

[0064] • Support for third-party AI model libraries.

[0065] The support status of third-party AI model libraries indicates the AI ​​models supported by the device. For example, the AI ​​models supported by the device may include at least one of the following: Convolutional Neural Network (CNN) models, Deep Neural Network (DNN) models, Recurrent Neural Network (RNN) models, and Transformers models.

[0066] Among them, the third-party AI model library is used to provide the terminal with AI-based CSI processing capabilities. That is, if the terminal supports the third-party AI model library, the terminal can perform AI-based CSI processing through the AI ​​models provided in the third-party AI model library.

[0067] For example, the communication protocol predefines at least one AI model; if the support of a third-party AI model library indicates that the third-party AI model library supports the terminal's invocation of any one of the at least one AI model, the network device can determine that the terminal supports AI-based CSI processing, that is, the terminal has AI-based CSI processing capability.

[0068] For example, the support status of third-party AI model libraries can be reported by the terminal through a bitmap.

[0069] Step 220: Determine the terminal's AI-based CSI processing capability based on the above static capabilities.

[0070] The terminal determines its AI-based CSI processing capability based on its AI-based static capabilities. Optionally, the AI-based CSI processing capability includes the minimum latency of AI-based CSI processing; the minimum latency of AI-based CSI processing of the terminal corresponding to the AI-based static capabilities is determined from the correspondence relationship; wherein, the correspondence relationship refers to the mapping relationship between the AI-based static capabilities and the minimum latency of AI-based CSI processing.

[0071] For example, the above-mentioned correspondence is set in the network device, or the above-mentioned correspondence is defined in the communication protocol, and the above-mentioned correspondence includes at least one of the following:

[0072] The mapping relationship between hardware information with AI processing capabilities and the minimum latency of AI-based CSI processing;

[0073] The mapping relationship between the support status of AI processing platforms and the minimum latency of AI-based CSI processing;

[0074] The mapping relationship between the support of third-party AI model libraries and the minimum latency of AI-based CSI processing;

[0075] The mapping relationship between hardware information with AI processing capabilities, the support status of AI processing platforms, and the minimum latency of AI-based CSI processing;

[0076] The mapping relationship between hardware information with AI processing capabilities, support for third-party AI model libraries, and the minimum latency of AI-based CSI processing;

[0077] The mapping relationship between the availability of AI processing platform support, the availability of third-party AI model libraries, and the minimum latency of AI-based CSI processing.

[0078] For example, after obtaining the terminal's AI-based CSI processing capability, the network device configures the AI-based CSI processing method and the allowable latency for AI-based CSI processing for the terminal based on the terminal's AI-based CSI processing capability.

[0079] It should be noted that if the terminal does not have its own AI processing capabilities, it will not report AI-based static capabilities. In this case, the network device defaults to the terminal only supporting CSI processing methods other than AI-based CSI processing methods, and the terminal reports based on the aforementioned other CSI processing methods.

[0080] In summary, the AI-based CSI processing capability determination method provided in this embodiment allows network devices to determine the terminal's AI-based CSI processing capability based on the terminal's reported AI-based static capabilities. This method provides a capability reporting mechanism for the terminal's own CSI processing capability, which supports network devices in judging the terminal's CSI processing capability, thereby configuring AI-based CSI compression that matches the terminal's CSI processing capability.

[0081] The network device's AI-based CSI processing capability for the terminal can be determined directly based on the correspondence between AI-based static capabilities and AI-based CSI processing capabilities, such as... Figure 2 As shown in step 220; it can also be determined based on the AI ​​processing speed reported by the terminal, for example, such as Figure 3 As shown, the AI-based CSI processing capability determination method includes the following steps:

[0082] Step 310: Receive the reported AI processing speed from the terminal.

[0083] The AI ​​processing speed of the terminal is measured by the terminal and reported to the network device. Alternatively, the AI ​​processing speed of the terminal is indicated by the hardware parameters on the terminal, which the terminal obtains and reports to the network device; for example, the hardware parameters of the AI ​​chip on the terminal indicate the AI ​​processing speed of the AI ​​chip, which the terminal obtains from a designated storage location and then reports to the network device.

[0084] Optionally, the aforementioned AI processing speed is actively reported by the terminal to the network device. Alternatively, the aforementioned AI processing speed is requested to be reported by the terminal from the network device; for example, before receiving the reported AI processing speed from the terminal, the network device sends a request to the terminal to obtain the AI ​​processing speed, and the terminal reports its own AI processing speed to the network device based on the aforementioned request.

[0085] Step 320: Determine the terminal's AI-based CSI processing capability based on the AI ​​processing speed and the model information of the AI ​​model used to process CSI.

[0086] The model information of the AI ​​model used for CSI processing mentioned above is configured by the network device for the terminal; for example, the model information of the AI ​​model used for CSI processing mentioned above can be configured by the network device for the terminal based on the terminal's static AI capabilities. For example, the model information mentioned above may include the model size. The model information of the AI ​​model used for CSI processing mentioned above is the model information of the switched AI model configured by the network device for the terminal.

[0087] For example, after obtaining the terminal's AI-based CSI processing capability, the network device configures the AI-based CSI processing method and the allowable latency for AI-based CSI processing for the terminal based on the terminal's AI-based CSI processing capability.

[0088] For example, the aforementioned AI-based CSI processing capability can be represented by an AI-based CSI processing capability level.

[0089] In summary, the AI-based CSI processing capability determination method provided in this embodiment allows network devices to determine the terminal's AI-based CSI processing capability based on the terminal's AI processing speed and the model information of the AI ​​model used for CSI processing. When the terminal has AI processing capability, the method also supports the network device in determining whether the CSI processing capability of the AI ​​model configured for the terminal meets the latency requirements, and if so, configuring the terminal to perform CSI compression based on AI.

[0090] In other embodiments, the network device may also receive AI-based CSI processing capabilities reported by the terminal, for example, such as Figure 4 As shown, the AI-based CSI processing capability determination method may further include the following steps:

[0091] Step 410: Send model information of the AI ​​model used to process CSI to the terminal.

[0092] The network device sends the model information of the AI ​​model used to process CSI to the terminal. The terminal then determines its own AI-based CSI processing capability based on its own AI processing speed and the model information of the AI ​​model used to process CSI.

[0093] For example, the network device configures model information for an AI model used to process CSI on the terminal based on the terminal's static AI-based capabilities, and sends the aforementioned model information for the AI ​​model used to process CSI to the terminal. For instance, the model information may include the model size. The aforementioned model information for the AI ​​model used to process CSI may be the model information of a switched AI model configured by the network device for the terminal.

[0094] For example, the AI ​​processing speed of the terminal is measured by the terminal itself. Alternatively, the AI ​​processing speed of the terminal is indicated by the hardware parameters on the terminal and obtained by the terminal; for instance, the hardware parameters of the AI ​​chip on the terminal indicate the AI ​​processing speed of the AI ​​chip, and the terminal obtains the AI ​​processing speed from a specified storage location.

[0095] Step 420: Receive the terminal's AI-based CSI processing capability reported by the terminal. The AI-based CSI processing capability is determined by the terminal based on model information of the AI ​​model used for CSI processing.

[0096] For example, after obtaining the terminal's AI-based CSI processing capability, the network device configures the AI-based CSI processing method and the allowable latency for AI-based CSI processing for the terminal based on the terminal's AI-based CSI processing capability.

[0097] For example, the AI-based CSI processing capability described above can be represented by an AI-based CSI processing capability level.

[0098] In summary, the AI-based CSI processing capability determination method provided in this embodiment allows the network device to receive the terminal's AI-based CSI processing capability reported by the terminal after the network device sends the model information of the AI ​​model used for CSI processing. This method supports the network device in determining whether the CSI processing capability of the AI ​​model configured for the terminal meets the latency requirements when the terminal has AI processing capabilities. If the requirements are met, the terminal is configured to perform CSI compression based on AI.

[0099] Network devices, based on the terminal's AI-enabled CSI processing capabilities, configure CSI processing methods for the terminal, thereby enabling switching between CSI processing methods. For example, such as... Figure 5 The diagram illustrates a flowchart of a CSI processing mode switching method provided in an exemplary embodiment of this disclosure. This method is applied to... Figure 1 In the network device of the communication system shown, the method includes:

[0100] Step 510: Send a switching instruction to the terminal. The switching instruction is used to indicate that the first CSI processing mode is switched to the second CSI processing mode.

[0101] The first CSI processing method is an AI-based CSI processing method, and the second CSI processing method is another CSI processing method besides the AI-based CSI processing method; or, the second CSI processing method is an AI-based CSI processing method, and the first CSI processing method is another CSI processing method besides the AI-based CSI processing method. For example, the aforementioned other CSI processing methods can be traditional CSI processing methods. Optionally, the latency corresponding to the AI-based CSI processing method is less than the latency corresponding to the aforementioned other CSI processing methods.

[0102] Optionally, the first CSI processing method is an AI-based CSI processing method, and the second CSI processing method is a CSI processing method other than the AI-based CSI processing method; when the network device sends a handover command to the terminal when the terminal feedback probability increases while using AI-based CSI compression.

[0103] Optionally, the aforementioned handover instructions include semi-static instructions. Semi-static instructions are used to instruct the terminal to provide CSI feedback according to the CSI processing method indicated by the handover instructions for a period of time. Specifically, when using AI-based CSI compression, the terminal's feedback refers to feedback on whether the data transmission was correct.

[0104] Optionally, the aforementioned handover instructions include dynamic instructions. Dynamic instructions are used to instruct the terminal to perform CSI feedback in accordance with the CSI processing method indicated by the handover instructions.

[0105] For example, when a network device triggers a semi-static switch of the CSI processing mode, the switching instruction is a semi-static instruction; when a network device triggers a dynamic switch of the CSI processing mode, the switching instruction is a dynamic instruction.

[0106] In summary, the CSI processing mode switching method provided in this embodiment supports both semi-static and dynamic switching between two CSI processing modes triggered by network devices. Since semi-static commands consume fewer signaling resources, network devices can save signaling resources by using semi-static commands to instruct terminals to switch CSI processing modes. Dynamic commands, on the other hand, can more flexibly implement switching between CSI processing modes. Switching CSI processing modes allows CSI compression to better match the latency requirements of terminal services and the transmission speed of the channel environment.

[0107] In some embodiments, the switching of the CSI processing method can also be a semi-static switching triggered by the terminal. For example, when the terminal's AI processing capabilities and latency requirements do not match, the terminal sends a switching request to the network device. The switching request is used to request a switch of the CSI processing method. The network device sends a switching instruction to the terminal. The terminal switches from the first CSI processing method to the second CSI processing method based on the switching instruction. That is, the method also supports semi-static switching between two CSI processing methods triggered by the terminal, thereby achieving the effect of saving signaling resources.

[0108] The switching of the aforementioned CSI processing method can also be triggered by the terminal, for example, such as... Figure 6 The diagram illustrates a flowchart of a CSI processing mode switching method provided in another exemplary embodiment of this disclosure, which is applied to... Figure 1 In the network device of the communication system shown, the method includes:

[0109] Step 610: Receive instruction information reported by the terminal. The instruction information is used to indicate switching from the first CSI processing mode to the second CSI processing mode.

[0110] The first CSI processing method is an AI-based CSI processing method, and the second CSI processing method is another CSI processing method besides the AI-based CSI processing method; or, the second CSI processing method is an AI-based CSI processing method, and the first CSI processing method is another CSI processing method besides the AI-based CSI processing method. For example, the aforementioned other CSI processing methods can be traditional CSI processing methods. Optionally, the latency corresponding to the AI-based CSI processing method is less than the latency corresponding to the aforementioned other CSI processing methods.

[0111] The aforementioned instruction information is used to instruct the network device to process the terminal's feedback information regarding CSI based on the second CSI processing method. For example, the aforementioned instruction information is reported to the network device by the terminal after switching from the first CSI processing method to the second CSI processing method. For instance, the aforementioned instruction information is reported to the network device by the terminal after switching from the first CSI processing method to the second CSI processing method based on the degree of matching between the latency required by the service and the latency corresponding to the AI ​​processing capability; or, the aforementioned instruction information is reported to the network device by the terminal after switching from the first CSI processing method to the second CSI processing method based on the degree of matching between the transmission rate corresponding to the channel environment and the latency corresponding to the AI ​​processing capability. Wherein, the aforementioned degree of matching is used to indicate whether the latency required by the service matches or does not match the latency corresponding to the AI ​​processing capability; or, the aforementioned degree of matching is used to indicate whether the transmission rate corresponding to the channel environment matches or does not match the latency corresponding to the AI ​​processing capability.

[0112] For example, a network device can receive indication information sent by a terminal via a notification command, and then also receive CSI feedback information sent by the terminal. That is, the terminal can send indication information to the network device via a notification command; after sending the notification command to the network device, it also sends CSI feedback information to the network device.

[0113] For example, the network device can also receive CSI feedback information sent by the terminal, the CSI feedback information carrying the aforementioned indication information of the second CSI processing method after the terminal switches; the network device obtains the indication information from the CSI feedback information; and processes the terminal's CSI feedback information according to the indication information, based on the second CSI processing method. That is, the terminal can carry the indication information of the CSI processing method through the CSI feedback information.

[0114] For example, the indication information is reported to the network device by the terminal after switching from the first CSI processing method to the second CSI processing method when there is a mismatch between its AI processing capability and latency requirements. For example, the latency requirements include the latency required by the terminal's services. The matching of the latency corresponding to the AI ​​processing capability with the latency required by the terminal's services includes at least one of the following:

[0115] The latency corresponding to AI processing capabilities is the same as the latency required by terminal services;

[0116] The latency corresponding to the AI ​​processing capability is within the latency range required by the terminal business.

[0117] The mismatch between the latency required by AI processing capabilities and the latency required by terminal services includes at least one of the following:

[0118] The latency required by AI processing capabilities is greater than the maximum latency required by terminal services.

[0119] The latency corresponding to AI processing capabilities is less than the minimum latency required by terminal services.

[0120] When the terminal uses the first CSI processing method, the latency corresponding to the AI ​​processing capability on the terminal is less than the minimum latency required by the terminal service. In other words, the AI ​​processing capability on the terminal does not match the latency required by the terminal service. The terminal switches the first CSI processing method to the second CSI processing method and sends an indication message to the network device.

[0121] In summary, the CSI processing mode switching method provided in this embodiment supports dynamic switching of CSI processing modes triggered by the terminal, enabling flexible switching of CSI processing modes; and the terminal-triggered switching speeds up the switching efficiency of CSI processing modes.

[0122] Figure 7 A flowchart of an AI-based CSI processing capability determination method provided in an exemplary embodiment of this disclosure is shown, which is applied to... Figure 1 In the terminal of the communication system shown, the method includes:

[0123] Step 710: Report the terminal's AI-based static capabilities to the network device. The AI-based static capabilities are used to determine the AI-based CSI processing capabilities.

[0124] Optionally, the AI-based static capabilities include at least one of the following:

[0125] • Hardware information with AI processing capabilities;

[0126] The hardware information indicating AI processing capabilities is used to indicate the AI ​​processing capabilities of the terminal. For example, the hardware information indicating AI processing capabilities includes hardware information used by the terminal to process CSI compression; for instance, the hardware information used by the terminal to process CSI compression includes at least one of the following: information about the AI ​​processing chip and information about the graphics processor.

[0127] Hardware information with AI processing capabilities can also be used to indicate the level of AI processing capabilities possessed by the terminal. For example, the hardware information used by the terminal to process CSI compression includes the hardware capability level. For instance, the hardware capability level can be defined based on the hardware's performance parameters, that is, the hardware level can be defined based on the hardware capabilities; the terminal determines and reports the hardware capability level based on the hardware's performance parameters.

[0128] Alternatively, the hardware information used by the terminal to process CSI compression includes hardware performance parameters, such as the performance parameters of the AI ​​processing chip and the graphics processing unit (GPU). Then, the network device can directly use these hardware performance parameters to determine the terminal's AI-based CSI processing capability; or, the network device can determine the hardware capability level based on the hardware performance parameters, and then determine the terminal's AI-based CSI processing capability based on the hardware capability level. That is, the information about the AI ​​processing chip can include its performance parameters, and the information about the graphics processing unit can include its performance parameters.

[0129] • Support status of the AI ​​processing platform;

[0130] The AI ​​processing platform support information indicates which AI processing platforms the terminal supports. Specifically, the AI ​​processing platform provides the terminal with AI-based CSI processing capabilities; that is, if the terminal supports an AI processing platform, it can perform AI-based CSI processing through that platform.

[0131] For example, AI processing platforms include TensorFlow, an end-to-end open-source machine learning platform.

[0132] For example, the communication protocol predefines at least one AI processing platform / framework; if the support status of the AI ​​processing platform indicates that the terminal supports any one of the at least one AI processing platform / framework, the network device determines that the terminal supports AI-based CSI processing, that is, the terminal has AI-based CSI processing capability.

[0133] For example, the terminal reports the support status of the AI ​​processing platform through a bitmap.

[0134] • Support for third-party AI model libraries.

[0135] The support information for third-party AI model libraries indicates the AI ​​models supported by the device. For example, the AI ​​models supported by the device may include at least one of CNN models, DNN models, RNN models, and Transformers models.

[0136] Among them, the third-party AI model library is used to provide the terminal with AI-based CSI processing capabilities. That is, if the terminal supports the third-party AI model library, the terminal can perform AI-based CSI processing through the AI ​​models provided in the third-party AI model library.

[0137] For example, the communication protocol predefines at least one AI model; if the support of a third-party AI model library indicates that the third-party AI model library supports the terminal's invocation of any one of the at least one AI model, the network device determines that the terminal supports AI-based CSI processing, that is, the terminal has AI-based CSI processing capability.

[0138] For example, the terminal reports the support status of third-party AI model libraries through a bitmap.

[0139] Optionally, the AI-based CSI processing capability includes the minimum latency of AI-based CSI processing; the AI-based static capability is used to determine the minimum latency of the terminal's AI-based CSI processing from the correspondence; wherein, the correspondence refers to the mapping relationship between the AI-based static capability and the minimum latency of AI-based CSI processing.

[0140] In some embodiments, the terminal also reports its AI processing speed to the network device. The AI ​​processing speed is used, together with model information of the AI ​​model used to process CSI, to determine the terminal's AI-based CSI processing capability.

[0141] In other embodiments, the terminal also reports its own AI-based CSI processing capabilities. For example, the terminal receives model information of an AI model for processing CSI sent by a network device; determines the terminal's AI-based CSI processing capabilities based on the model information; and reports the terminal's AI-based CSI processing capabilities to the network device.

[0142] For example, the terminal also receives from the network device the allowed latency of AI-based CSI processing configured for the terminal based on the terminal's AI-based CSI processing capabilities.

[0143] In summary, the AI-based CSI processing capability determination method provided in this embodiment determines the terminal's AI-based CSI processing capability by having the network device determine the terminal's AI-based static capabilities based on the terminal's reported AI-based capabilities. This method provides a reporting mechanism for the terminal's own CSI processing capabilities, which supports the network device in judging the terminal's capabilities, thereby configuring AI-based CSI compression that matches the terminal's capabilities.

[0144] Network devices configure CSI processing methods for terminals based on the terminal's AI-powered CSI processing capabilities. The terminals then switch between CSI processing methods based on the network device's configuration. For example,... Figure 8 As shown, the method for switching CSI processing modes is as follows:

[0145] Step 810: Receive a switching instruction sent by the network device. The switching instruction is used to indicate that the first CSI processing mode is switched to the second CSI processing mode.

[0146] The first CSI processing method is an AI-based CSI processing method, and the second CSI processing method is another CSI processing method besides the AI-based CSI processing method; or, the second CSI processing method is an AI-based CSI processing method, and the first CSI processing method is another CSI processing method besides the AI-based CSI processing method. For example, the aforementioned other CSI processing methods can be traditional CSI processing methods. Optionally, the latency corresponding to the AI-based CSI processing method is less than the latency corresponding to the aforementioned other CSI processing methods.

[0147] Optionally, when the first CSI processing method is an AI-based CSI processing method and the second CSI processing method is another CSI processing method besides the AI-based CSI processing method, the handover instruction is sent to the terminal by the network device when the terminal's feedback probability increases when using AI-based CSI compression.

[0148] Optionally, the aforementioned handover instructions include semi-static instructions. Semi-static instructions are used to instruct the terminal to provide CSI feedback according to the CSI processing method indicated by the handover instructions for a period of time. Specifically, when using AI-based CSI compression, the terminal's feedback refers to feedback on whether the data transmission was correct.

[0149] Optionally, the aforementioned handover instructions include dynamic instructions. Dynamic instructions are used to instruct the terminal to perform CSI feedback in accordance with the CSI processing method indicated by the handover instructions.

[0150] For example, when a network device triggers a semi-static switch of the CSI processing mode, the switching instruction is a semi-static instruction; when a network device triggers a dynamic switch of the CSI processing mode, the switching instruction is a dynamic instruction.

[0151] Step 820: Switch from the first CSI processing mode to the second CSI processing mode.

[0152] In summary, the CSI processing mode switching method provided in this embodiment supports both semi-static and dynamic switching between two CSI processing modes triggered by network devices. Since semi-static commands consume fewer signaling resources, network devices can save signaling resources by using semi-static commands to instruct terminals to switch CSI processing modes. Dynamic commands, on the other hand, can more flexibly implement switching between CSI processing modes. Switching CSI processing modes allows CSI compression to better match the latency requirements of terminal services and the transmission speed of the channel environment.

[0153] The terminal can also notify the network device after switching the CSI processing mode, for example, such as Figure 9 As shown, the method for switching CSI processing modes is as follows:

[0154] Step 910: Switch the first CSI processing mode to the second CSI processing mode.

[0155] The first CSI processing method is an AI-based CSI processing method, and the second CSI processing method is another CSI processing method besides the AI-based CSI processing method; or, the second CSI processing method is an AI-based CSI processing method, and the first CSI processing method is another CSI processing method besides the AI-based CSI processing method. For example, the aforementioned other CSI processing methods can be traditional CSI processing methods. Optionally, the latency corresponding to the AI-based CSI processing method is less than the latency corresponding to the aforementioned other CSI processing methods.

[0156] For example, the terminal switches from the first CSI processing mode to the second CSI processing mode based on the degree of matching between the latency required by the service and the latency corresponding to the AI ​​processing capability; or, the terminal switches from the first CSI processing mode to the second CSI processing mode based on the degree of matching between the transmission rate corresponding to the channel environment and the latency corresponding to the AI ​​processing capability. Here, the aforementioned degree of matching is used to indicate whether the latency required by the service and the latency corresponding to the AI ​​processing capability match or do not match; or, the aforementioned degree of matching is used to indicate whether the transmission rate corresponding to the channel environment and the latency corresponding to the AI ​​processing capability match or do not match.

[0157] For example, if there is a mismatch between the terminal's AI processing capabilities and latency requirements, it may switch from the first CSI processing method to the second CSI processing method. For instance, latency requirements include the latency required by the terminal's services. The mismatch between the latency corresponding to the AI ​​processing capabilities and the latency required by the terminal's services includes at least one of the following:

[0158] The latency required by AI processing capabilities is greater than the maximum latency required by terminal services.

[0159] The latency corresponding to AI processing capabilities is less than the minimum latency required by terminal services.

[0160] When the first CSI processing method is used on the terminal, the latency corresponding to the AI ​​processing capability on the terminal is less than the minimum latency required by the terminal business. In other words, the AI ​​processing capability on the terminal does not match the latency required by the terminal business, so the terminal switches the first CSI processing method to the second CSI processing method.

[0161] Step 920: Send an instruction message to the network device, the instruction message being used to instruct the switching of the first CSI processing mode to the second CSI processing mode.

[0162] The first CSI processing method is an AI-based CSI processing method, and the second CSI processing method is a CSI processing method other than the AI-based CSI processing method; or, the second CSI processing method is an AI-based CSI processing method, and the first CSI processing method is a CSI processing method other than the AI-based CSI processing method.

[0163] In summary, the CSI processing mode switching method provided in this embodiment supports dynamic switching of CSI processing modes triggered by the terminal, enabling flexible switching of CSI processing modes; and the terminal-triggered switching speeds up the switching efficiency of CSI processing modes.

[0164] For example, the AI-based CSI processing capability determination method in the above embodiments mainly includes:

[0165] Key Point 1: Static capability reporting based on AI.

[0166] When the terminal has AI processing capabilities, it reports at least one of the following capabilities to the network device:

[0167] • Hardware information indicating AI processing capabilities, including whether the terminal contains hardware for CSI compression processing, such as whether it has an AI processing chip or a GPU. The terminal can report whether it supports AI processing capabilities based on this information. Furthermore, the terminal can define hardware capability levels according to their level of capability and report these levels.

[0168] • AI processing platform support, such as whether it supports TensorFlow. Specifically, the communication protocol can predefine at least one general AI processing platform / framework. The terminal can report information via Bitmap.

[0169] • Support for third-party AI model libraries, such as whether they support CNN, DNN, RNN, and Transform models. Specifically, the communication protocol can predefine at least one common AI model. The terminal can report information via Bitmap.

[0170] When the terminal lacks AI processing capabilities, it does not report. In this case, the network device assumes that the terminal only supports traditional CSI processing and reporting methods.

[0171] Key Point Two: Determining CSI processing capabilities.

[0172] When network devices determine that a terminal has AI processing capabilities, it is necessary to further determine whether the AI's CSI processing can meet latency requirements.

[0173] 1) Network devices directly determine CSI processing capabilities based on the AI-based static capabilities reported by the terminal. For example, if the protocol predefines the mapping relationship between AI-based static capabilities and CSI processing latency, the network device can determine the minimum latency for CSI processing based on the reported AI-based static capabilities.

[0174] 2) The terminal further reports its AI processing speed and other parameters. The network device determines the terminal's CSI processing capability level based on the AI ​​processing speed reported by the terminal and the model size of the AI ​​model that the network device will use. The AI ​​processing speed and other parameters reported by the terminal can be reported proactively by the terminal or based on a request from the network device.

[0175] 3) The network device sends the model information of the AI ​​model to the terminal for CSI processing. The model information may include model size information or specific model information. The terminal judges the CSI processing capability of the AI ​​model and finally reports the CSI processing capability level to the network device.

[0176] The network device determines whether to configure AI-based CSI processing for the terminal and the allowable processing latency based on the CSI processing capability level fed back by the terminal.

[0177] Key Point 3: Switching between CSI processing methods.

[0178] For terminals with AI processing capabilities, it is possible to switch between AI-based CSI processing and traditional CSI processing.

[0179] 1) Network device-triggered handover. For example, when a network device detects that the probability of a terminal responding with NACK increases when using AI-based CSI compression, the network device can then configure the terminal to switch to the traditional CSI processing method (i.e., CSI feedback method).

[0180] 2) Terminal-triggered switching. For example, a terminal can request the network to switch to the traditional CSI processing method based on the current AI processing load and the degree of matching between the latency requirements when using AI processing.

[0181] In summary, the AI-based CSI processing capability determination method provided in this embodiment allows network devices to determine a terminal's AI-based CSI processing capability based on its reported static AI-based capabilities. This method provides a mechanism for terminals to report their own CSI processing capabilities, supporting network devices in judging terminal capabilities and configuring AI-based CSI compression that matches the terminal's capabilities. Furthermore, it also enables switching between CSI processing methods based on AI processing capabilities.

[0182] Figure 10 The diagram illustrates a block diagram of an AI-based CSI processing capability determination apparatus provided in an exemplary embodiment of this disclosure. This apparatus can be implemented as part or all of a network device through software, hardware, or a combination of both. The apparatus includes:

[0183] The first receiving module 1010 is configured to receive the reported AI-based static capabilities of the terminal.

[0184] The first processing module 1020 is configured to determine the AI-based CSI processing capability of the terminal based on the static capabilities.

[0185] In some embodiments, the AI-based CSI processing capability includes the minimum latency of AI-based CSI processing;

[0186] The first processing module 1020 is configured to determine the minimum latency of AI-based CSI processing of the terminal corresponding to the static capability from the correspondence relationship; wherein the correspondence relationship refers to the mapping relationship between the static capability and the minimum latency of AI-based CSI processing.

[0187] In some embodiments, the static capability includes at least one of the following:

[0188] Hardware information with AI processing capabilities;

[0189] Support status of AI processing platforms;

[0190] Support for third-party AI model libraries.

[0191] In some embodiments,

[0192] The first receiving module 1010 is configured to receive the reported AI processing speed of the terminal;

[0193] The first processing module 1020 is configured to determine the AI-based CSI processing capability of the terminal based on the AI ​​processing speed and the model information of the AI ​​model used to process CSI.

[0194] In some embodiments, the AI ​​processing speed is reported by the terminal to the network device proactively, or the AI ​​processing speed is reported by the terminal at the request of the network device.

[0195] In some embodiments, the apparatus further includes:

[0196] The first sending module 1030 is configured to send model information for processing CSI AI models to the terminal;

[0197] The first receiving module 1010 is configured to receive the AI-based CSI processing capability of the terminal reported by the terminal, wherein the AI-based CSI processing capability is determined by the terminal based on the model information.

[0198] In some embodiments, the first processing module 1020 is configured to configure the allowable latency for AI-based CSI processing for the terminal based on the AI-based CSI processing capability.

[0199] In some embodiments, the apparatus further includes:

[0200] The first sending module 1030 is configured to send a switching instruction to the terminal when the feedback probability of the terminal increases when using AI-based CSI compression. The switching instruction is used to indicate switching from the first CSI processing mode to the second CSI processing mode.

[0201] The first CSI processing method is an AI-based CSI processing method, and the second CSI processing method is a CSI processing method other than the AI-based CSI processing method.

[0202] In some embodiments, the first receiving module 1010 is configured to receive indication information reported by the terminal, the indication information being used to indicate switching from a first CSI processing mode to a second CSI processing mode;

[0203] The first CSI processing method is an AI-based CSI processing method, and the second CSI processing method is a CSI processing method other than the AI-based CSI processing method.

[0204] In some embodiments, the indication information is reported to the network device by the terminal after switching from the first CSI processing mode to the second CSI processing mode when there is a mismatch between its own AI processing capabilities and latency requirements.

[0205] Figure 11 A block diagram of an AI-based CSI processing capability determination apparatus provided in another exemplary embodiment of this disclosure is shown. This apparatus can be implemented as part or all of a terminal through software, hardware, or a combination of both. The apparatus includes:

[0206] The second sending module 1110 is configured to report the terminal's AI-based static capabilities to the network device, the AI-based static capabilities being used to determine the AI-based CSI processing capabilities.

[0207] In some embodiments, the AI-based CSI processing capability includes the minimum latency of AI-based CSI processing;

[0208] The AI-based static capability is used to determine the minimum latency of the AI-based CSI processing of the terminal from the correspondence; wherein, the correspondence refers to the mapping relationship between the static capability and the minimum latency of the AI-based CSI processing.

[0209] In some embodiments, the static capability includes at least one of the following:

[0210] Hardware information with AI processing capabilities;

[0211] Support status of AI processing platforms;

[0212] Support for third-party AI model libraries.

[0213] In some embodiments, the second sending module 1110 is configured to report the AI ​​processing speed of the terminal to the network device, the AI ​​processing speed being used together with model information of the AI ​​model used for processing CSI to determine the terminal's AI-based CSI processing capability.

[0214] In some embodiments, the apparatus further includes:

[0215] The second receiving module 1120 is configured to receive model information of the AI ​​model for processing CSI sent by the network device;

[0216] The second processing module 1130 is configured to determine the AI-based CSI processing capability of the terminal based on the model information.

[0217] The second sending module 1110 is configured to report the terminal's AI-based CSI processing capabilities to the network device.

[0218] In some embodiments, the apparatus further includes:

[0219] The second receiving module 1120 is configured to receive the allowable latency of AI-based CSI processing configured for the terminal by the network device based on the AI-based CSI processing capability of the terminal.

[0220] In some embodiments, the apparatus further includes:

[0221] The second receiving module 1120 is configured to receive a switching instruction sent by the network device. The switching instruction is used to indicate switching from the first CSI processing mode to the second CSI processing mode. The switching instruction is sent by the network device to the terminal when the feedback probability of the terminal increases when using AI-based CSI compression.

[0222] The first CSI processing method is an AI-based CSI processing method, and the second CSI processing method is a CSI processing method other than the AI-based CSI processing method.

[0223] In some embodiments, the second sending module 1110 is configured to send indication information to the network device, the indication information being used to indicate switching from the first CSI processing mode to the second CSI processing mode;

[0224] The first CSI processing method is an AI-based CSI processing method, and the second CSI processing method is a CSI processing method other than the AI-based CSI processing method.

[0225] In some embodiments, the apparatus further includes:

[0226] The second processing module 1130 is configured to switch the first CSI processing mode to the second CSI processing mode before sending the instruction information to the network device, in the event of a mismatch between the AI ​​processing capability and latency requirements of the terminal.

[0227] Figure 12 A schematic diagram of the structure of a UE provided in an exemplary embodiment of the present disclosure is shown. The UE includes: a processor 1201, a receiver 1202, a transmitter 1203, a memory 1204, and a bus 1205.

[0228] The processor 1201 includes one or more processing cores. The processor 1201 executes various functional applications and information processing by running software programs and modules.

[0229] The receiver 1202 and the transmitter 1203 can be implemented as a communication component, which can be a communication chip.

[0230] The memory 1204 is connected to the processor 1201 via the bus 1205.

[0231] The memory 1204 can be used to store at least one instruction, and the processor 1201 is used to execute the at least one instruction to implement the various steps in the above method embodiments.

[0232] Furthermore, the memory 1204 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, including but not limited to: magnetic disks or optical disks, electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), static random-access memory (SRAM), read-only memory (ROM), magnetic storage, flash memory, and programmable read-only memory (PROM).

[0233] In an exemplary embodiment, a non-transitory computer-readable storage medium including instructions is also provided, such as a memory including instructions that can be executed by the processor of the UE to complete the aforementioned AI-based CSI processing capability determination method. For example, the non-transitory computer-readable storage medium may be a ROM, random-access memory (RAM), a compact disc read-only memory (CD-ROM), magnetic tape, floppy disk, and optical data storage device, etc.

[0234] A non-transitory computer-readable storage medium, wherein when instructions in the non-transitory computer storage medium are executed by a processor of a UE, the user equipment (UE) is able to execute the aforementioned AI-based CSI processing capability determination method.

[0235] Figure 13 This is a block diagram illustrating a network device 1300 according to an exemplary embodiment. The network device 1300 may be a base station.

[0236] Network device 1300 may include: processor 1301, receiver 1302, transmitter 1303, and memory 1304. Receiver 1302, transmitter 1303, and memory 1304 are respectively connected to processor 1301 via a bus.

[0237] The processor 1301 includes one or more processing cores. The processor 1301 executes the AI-based CSI processing capability determination method provided in this embodiment by running software programs and modules. The memory 1304 can be used to store software programs and modules. Specifically, the memory 1304 can store an operating system 13041 and at least one application module 13042 required for a given function. The receiver 1302 is used to receive communication data sent by other devices, and the transmitter 1303 is used to send communication data to other devices.

[0238] An exemplary embodiment of this disclosure also provides a computer-readable storage medium storing at least one instruction, at least one program, a code set, or an instruction set, wherein the at least one instruction, the at least one program, the code set, or the instruction set is loaded and executed by the processor to implement the AI-based CSI processing capability determination method provided in the above-described method embodiments.

[0239] An exemplary embodiment of this disclosure also provides a computer program product including computer instructions stored in a computer-readable storage medium; a processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the AI-based CSI processing capability determination method provided in the various method embodiments described above.

[0240] It should be understood that "multiple" as used in this article refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone. The character " / " generally indicates that the preceding and following related objects have an "or" relationship.

[0241] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This disclosure is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the following claims.

[0242] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims.

Claims

1. A method for determining CSI processing capability based on AI, characterized in that, The method is performed by a network device, and the method includes: The static capabilities of the receiving terminal based on artificial intelligence (AI); The terminal's AI-based Channel State Indication (CSI) processing capability is determined based on the static capability, and the AI-based CSI processing capability includes the minimum latency of AI-based CSI processing. Determining the AI-based CSI processing capability of the terminal based on the static capabilities includes: The minimum latency of AI-based CSI processing for the terminal corresponding to the static capability is determined from the correspondence relationship; wherein the correspondence relationship refers to the mapping relationship between the static capability and the minimum latency of AI-based CSI processing.

2. The method according to claim 1, characterized in that, The static capabilities include at least one of the following: Hardware information with AI processing capabilities; Support status of AI processing platforms; Support for third-party AI model libraries.

3. The method according to claim 1, characterized in that, The method further includes: Receive the reported AI processing speed of the terminal; Based on the AI ​​processing speed and the model information of the AI ​​model used to process CSI, the AI-based CSI processing capability of the terminal is determined.

4. The method according to claim 3, characterized in that, The AI ​​processing speed is either actively reported by the terminal to the network device, or the AI ​​processing speed is reported by the terminal at the request of the network device.

5. The method according to claim 1, characterized in that, The method further includes: Send model information for the AI ​​model used to process CSI to the terminal; The terminal receives the AI-based CSI processing capability reported by the terminal, which is determined by the terminal based on the model information.

6. The method according to claim 5, characterized in that, The method further includes: Configure the terminal with the allowable latency for AI-based CSI processing based on the AI-based CSI processing capability.

7. The method according to claim 1, characterized in that, The method further includes: When the feedback probability of the terminal increases when using AI-based CSI compression, a switching instruction is sent to the terminal, the switching instruction being used to indicate switching from the first CSI processing mode to the second CSI processing mode; The first CSI processing method is an AI-based CSI processing method, and the second CSI processing method is a CSI processing method other than the AI-based CSI processing method.

8. The method according to claim 1, characterized in that, The method further includes: The terminal reports an instruction message, which is used to instruct the user to switch from the first CSI processing mode to the second CSI processing mode. The first CSI processing method is an AI-based CSI processing method, and the second CSI processing method is a CSI processing method other than the AI-based CSI processing method.

9. The method according to claim 8, characterized in that, The indication information is reported to the network device by the terminal after switching from the first CSI processing mode to the second CSI processing mode when there is a mismatch between its own AI processing capability and latency requirements.

10. A method for determining CSI processing capability based on AI, characterized in that, The method is executed by a terminal, and the method includes: The terminal's AI-based static capabilities are reported to the network device. These AI-based static capabilities are used to determine the AI-based CSI processing capabilities, which include the minimum latency of AI-based CSI processing. The AI-based static capabilities are used to determine the minimum latency of the terminal's AI-based CSI processing from the correspondence relationship. The correspondence refers to the mapping relationship between the static capability and the minimum latency of the AI-based CSI processing.

11. The method according to claim 10, characterized in that, The static capabilities include at least one of the following: Hardware information with AI processing capabilities; Support status of AI processing platforms; Support for third-party AI model libraries.

12. The method according to claim 10, characterized in that, The method further includes: The terminal's AI processing speed is reported to the network device. This AI processing speed, together with model information of the AI ​​model used to process CSI, is used to determine the terminal's AI-based CSI processing capability.

13. The method according to claim 10, characterized in that, The method further includes: Receive model information from the network device for processing CSI AI models; The AI-based CSI processing capability of the terminal is determined based on the model information; The terminal's AI-based CSI processing capabilities are reported to the network device.

14. The method according to claim 13, characterized in that, The method further includes: The network device receives the allowed latency for AI-based CSI processing configured for the terminal based on the terminal's AI-based CSI processing capability.

15. The method according to claim 10, characterized in that, The method further includes: The network device receives a switching instruction sent by the network device. The switching instruction is used to indicate switching from the first CSI processing mode to the second CSI processing mode. The switching instruction is sent to the terminal by the network device when the feedback probability of the terminal increases when using AI-based CSI compression. The first CSI processing method is an AI-based CSI processing method, and the second CSI processing method is a CSI processing method other than the AI-based CSI processing method.

16. The method according to claim 10, characterized in that, The method further includes: Send an instruction message to the network device, the instruction message being used to instruct the switching of the first CSI processing mode to the second CSI processing mode; The first CSI processing method is an AI-based CSI processing method, and the second CSI processing method is a CSI processing method other than the AI-based CSI processing method.

17. The method according to claim 16, characterized in that, Before sending the indication information to the network device, the following steps are included: In the event of a mismatch between the AI ​​processing capability and latency requirements of the terminal, the first CSI processing method will be switched to the second CSI processing method.

18. An AI-based CSI processing capability determination device, characterized in that, The device includes: The first receiving module is configured to receive the reported AI-based static capabilities of the terminal. The first processing module is configured to determine the terminal's AI-based CSI processing capability based on the static capability, wherein the AI-based CSI processing capability includes the minimum latency of AI-based CSI processing. The first processing module is configured to determine the minimum latency of AI-based CSI processing for the terminal corresponding to the static capability from the correspondence relationship; The correspondence refers to the mapping relationship between the static capability and the minimum latency of the AI-based CSI processing.

19. An AI-based CSI processing capability determination device, characterized in that, The device includes: The second sending module is configured to report the terminal's AI-based static capabilities to the network device. The AI-based static capabilities are used to determine the AI-based CSI processing capabilities, which include the minimum latency of the AI-based CSI processing. The AI-based static capabilities are used to determine the minimum latency of the terminal's AI-based CSI processing from the correspondence relationship. The correspondence refers to the mapping relationship between the static capability and the minimum latency of the AI-based CSI processing.

20. A network device, characterized in that, The network device includes: processor; A transceiver connected to the processor; The processor is configured to load and execute executable instructions to implement the AI-based CSI processing capability determination method as described in any one of claims 1 to 9.

21. A terminal, characterized in that, The terminal includes: processor; A transceiver connected to the processor; The processor is configured to load and execute executable instructions to implement the AI-based CSI processing capability determination method as described in any one of claims 10 to 17.

22. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores at least one instruction, at least one program, code set, or instruction set, wherein the at least one instruction, the at least one program, the code set, or the instruction set is loaded and executed by a processor to implement the AI-based CSI processing capability determination method as described in any one of claims 1 to 9, or the AI-based CSI processing capability determination method as described in any one of claims 10 to 17.

23. A computer program product, characterized in that, The computer program product includes computer instructions stored in a computer-readable storage medium; a processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the AI-based CSI processing capability determination method as described in any one of claims 1 to 9, or the AI-based CSI processing capability determination method as described in any one of claims 10 to 17.

24. A chip, characterized in that, The chip includes programmable logic circuitry and / or program instructions, which, when the chip is running, are used to implement the AI-based CSI processing capability determination method as described in any one of claims 1 to 9, or the AI-based CSI processing capability determination method as described in any one of claims 10 to 17.

Images