Channel state information processing method and apparatus, and device and storage medium

Abstract

The present application relates to the technical field of communications. Provided are a channel state information processing method and apparatus, and a device and a storage medium. The method comprises: receiving first information sent by a network device, wherein the first information is used for triggering and / or activating channel state information (CSI) configuration information; and transmitting CSI to the network device, wherein the CSI is transmitted on the basis of a CSI processing unit (CPU), a target resource and the CSI configuration information, and the target resource is associated with AI processing of the CSI. By means of the method, a failure in CSI calculation can be avoided, the success rate of CSI calculation is increased, and the success rate and accuracy of CSI transmission are improved.

Description

Channel state information processing methods, apparatus, equipment and storage media

[0001] This disclosure claims priority to Chinese Patent Application No. 202510064945.8, filed on January 15, 2025, entitled “Method, Apparatus, Device and Storage Medium for Channel State Information Processing”, the entire contents of which are incorporated herein by reference. Technical Field

[0002] This disclosure relates to the field of communication technology, specifically to a method, apparatus, device, and storage medium for processing channel state information. Background Technology

[0003] Channel State Information (CSI) describes the channel attributes of a communication link. Network devices can allocate communication resources reasonably and improve communication quality based on the CSI sent by the terminal devices.

[0004] In related technologies, terminal devices can calculate CSI based on CSI processing units (CPU). When the calculation time of multiple CSIs overlaps, the CPU required by the terminal device to calculate multiple CSIs cannot exceed the upper limit of the CPU in the terminal device.

[0005] However, future communication systems may also include CSI based on Artificial Intelligence (AI) processing. CPU generally represents the capability requirement for computing traditional CSI, which makes the accuracy of terminal equipment's assessment of the resources required to compute CSI low, and may lead to CSI computation failures, resulting in a low success rate of CSI transmission. Summary of the Invention

[0006] This disclosure provides a channel state information processing method, apparatus, device, and storage medium to solve the technical problems of CSI calculation and transmission, and its priority in related technologies.

[0007] In a first aspect, embodiments of this disclosure provide a channel state information processing method, applied to a terminal device, the channel state information processing method comprising:

[0008] Receive first information sent by the network device, the first information being used to trigger and / or activate Channel State Information (CSI) configuration information;

[0009] The CSI is sent to the network device. The CSI is sent based on the CSI processing unit CPU, the target resource, and the CSI configuration information. The target resource is associated with the AI ​​processing of the CSI.

[0010] In some embodiments, the target resource includes at least one of the following:

[0011] Computing resources used for AI data processing;

[0012] Memory resources used for AI data processing;

[0013] Cache resources used for AI data processing;

[0014] Storage space resources used for AI data processing.

[0015] In some embodiments, the type of AI data processing includes at least one of the following:

[0016] Based on the AI ​​data, perform inference and / or prediction using the AI ​​model;

[0017] Based on the AI ​​data, perform AI model training tasks or related data collection;

[0018] Monitor the performance of AI model inference and / or prediction.

[0019] In some embodiments, sending the CSI to the network device includes:

[0020] In the triggered and / or activated CSI configuration information, target CSI configuration information is determined, which is determined based on the target resources and / or CPU occupied by each CSI configuration information, and the target time for occupying the target resources and / or CPU.

[0021] The CSI is sent to the network device according to the target CSI configuration information.

[0022] In some embodiments, the target resources and / or CPU occupied by the CSI configuration information are determined based on at least one of the following:

[0023] The feedback quantity in the CSI configuration information;

[0024] High-level parameters associated with the CSI configuration information;

[0025] The parameters in the CSI configuration information;

[0026] The reference signal resources corresponding to the CSI configuration information;

[0027] Indicators or associated high-level parameters used to indicate the capabilities of terminal devices.

[0028] In some embodiments, the start time of the target time is determined based on at least one of the following:

[0029] The first symbol is the first symbol that triggers and / or activates the CSI configuration information, the last symbol, or the symbol following the last symbol.

[0030] The second symbol is the first symbol, the last symbol, or the symbol following the last symbol of the first reference signal transmission timing corresponding to the current CSI.

[0031] The third symbol is the first symbol, the last symbol, or the symbol following the last symbol of the last reference signal transmission timing corresponding to the current CSI transmission.

[0032] The fourth symbol is either the first symbol after the measurement at the last reference signal transmission timing is completed, or the first symbol after the measurement result is processed;

[0033] The fifth symbol is either the first symbol after completing AI model inference and / or prediction, or the Mth symbol after completing the measurement of the reference signal transmission timing, where M is associated with the terminal device capability.

[0034] In some embodiments, the end time of the target time is determined based on at least one of the following:

[0035] The fourth symbol;

[0036] The fifth symbol;

[0037] The sixth symbol is either the first or last symbol of the current CSI transmission.

[0038] The seventh symbol is either the first or last symbol corresponding to the CSI transmission period indicated by the CSI configuration information;

[0039] The eighth symbol is either the first or last symbol of the deactivation signaling corresponding to the CSI configuration information.

[0040] In some embodiments, if the CPU and / or the target resources are insufficient, the target CSI configuration information is determined according to the priority order of the CPU and / or the target resources and the CSI configuration information.

[0041] In some embodiments, the priority order of the CSI configuration information is determined according to at least one of the following:

[0042] The periodic information in the CSI configuration information;

[0043] The CSI configuration information specifies the triggering method for sending CSI.

[0044] The feedback quantity in the CSI configuration information;

[0045] Information in the CSI configuration information;

[0046] The number of reference signals corresponding to the CSI configuration information;

[0047] High-level parameter information associated with the CSI configuration information;

[0048] Serving the community;

[0049] The identifier of the CSI configuration information.

[0050] In some embodiments, determining the target CSI configuration information based on the priority order of the CPU and / or the target resource, and the CSI configuration information, includes:

[0051] According to the priority order, a CSI configuration information set is determined. The CSI configuration information in the CSI configuration information set is determined in descending order of priority. Furthermore, when calculating all the CSI configuration information in the CSI configuration information set, the CPU usage is less than the upper limit of the CPU, and the target resources used are less than the upper limit of the target resources.

[0052] The target CSI configuration information is determined from the CSI configuration information set.

[0053] In some embodiments, the method further includes:

[0054] Send the CPU limit and / or the target resource limit to the network device.

[0055] In some embodiments, the CSI transmission includes AI-based CSI transmission and non-AI-based CSI transmission, and the method further includes:

[0056] Determine whether the AI-based CSI transmission meets the handover conditions;

[0057] If the AI-based CSI transmission meets the switching conditions, the AI-based CSI transmission will be switched to the non-AI-based CSI transmission.

[0058] In some embodiments, the switching conditions include at least one of the following:

[0059] The number of AI-based CSI transmission failures is greater than a preset number;

[0060] The AI-based CSI transmission is not sent due to priority considerations;

[0061] The network device did not receive the AI-based CSI transmission.

[0062] In some embodiments, the handover conditions are predefined or determined by the network device, and whether the AI-based CSI transmission satisfies the handover conditions is determined by the terminal device or the network device.

[0063] In some embodiments, when the switching condition for the AI-based CSI transmission is determined by the terminal device, switching the AI-based CSI transmission to the non-AI-based CSI transmission includes:

[0064] Send a handover request to the network device;

[0065] The system receives a first indication message sent by a network device, the first indication message being used to indicate whether to switch the AI-based CSI transmission to the non-AI-based CSI transmission.

[0066] In some embodiments, when the switching condition for the AI-based CSI transmission is determined by the network device, switching the AI-based CSI transmission to the non-AI-based CSI transmission includes:

[0067] The system receives a second indication message sent by a network device, the second indication message being used to indicate switching the AI-based CSI transmission to the non-AI-based CSI transmission.

[0068] In some embodiments, the method further includes:

[0069] The system receives second information sent by a network device. The second information is used to activate and / or trigger first CSI configuration information and second CSI configuration information. The first CSI configuration information is used for AI-based CSI transmission, and the second CSI configuration information is used for performance monitoring of the first CSI configuration information.

[0070] In some embodiments, the second information includes an identifier of the first CSI configuration information, and the first CSI configuration information indicates the second CSI configuration information, or the first CSI configuration information and the second CSI configuration information are associated with a timeline;

[0071] The second information includes an identifier of the second CSI configuration information, and the second CSI configuration information indicates the first CSI configuration information, or the second CSI configuration information is associated with a timeline between the first CSI configuration information and the second CSI configuration information.

[0072] Secondly, embodiments of this disclosure provide a channel state information processing method, applied to a network device, the channel state information processing method comprising:

[0073] The first information sent to the terminal device is used to trigger and / or activate the Channel State Information (CSI) configuration information.

[0074] The terminal device receives a CSI sent by the terminal device. The CSI is sent based on the CPU, target resources, and CSI configuration information. The target resources are associated with the AI ​​processing of the CSI.

[0075] In some embodiments, the target resource includes at least one of the following:

[0076] Computing resources used for AI data processing;

[0077] Memory resources used for AI data processing;

[0078] Cache resources used for AI data processing;

[0079] Storage space resources used for AI data processing.

[0080] In some embodiments, the type of AI data processing includes at least one of the following:

[0081] Based on the AI ​​data, perform inference and / or prediction using the AI ​​model;

[0082] Based on the AI ​​data, perform AI model training tasks or related data collection;

[0083] Monitor the performance of AI model inference and / or prediction.

[0084] In some embodiments, receiving CSI sent by the terminal device includes:

[0085] According to the target CSI configuration information, the terminal device sends CSIs. The target CSI configuration information is determined based on the target resources and / or CPU occupied by each CSI configuration information, and the target time for occupying the target resources and / or CPU.

[0086] In some embodiments, the target resources and / or CPU occupied by the CSI configuration information are determined based on at least one of the following:

[0087] The feedback quantity in the CSI configuration information;

[0088] High-level parameters associated with the CSI configuration information;

[0089] The parameters in the CSI configuration information;

[0090] The reference signal resources corresponding to the CSI configuration information;

[0091] Indicators or associated high-level parameters used to indicate the capabilities of terminal devices.

[0092] In some embodiments, the start time of the target time is determined based on at least one of the following:

[0093] The first symbol is the first symbol, the last symbol, or the symbol following the last symbol after triggering and / or activating the CSI configuration information;

[0094] The second symbol is the first symbol, the last symbol, or the symbol following the last symbol of the first reference signal transmission timing corresponding to the current CSI.

[0095] The third symbol is the first symbol, the last symbol, or the symbol following the last symbol of the last reference signal transmission timing corresponding to the current CSI.

[0096] The fourth symbol is either the first symbol after the measurement at the last reference signal transmission timing is completed, or the first symbol after the measurement result is processed;

[0097] The fifth symbol is either the first symbol after completing AI model inference and / or prediction, or the Mth symbol after completing the measurement of the reference signal transmission timing, where M is associated with the capabilities of the terminal device.

[0098] In some embodiments, the end time of the target time is determined based on at least one of the following:

[0099] The fourth symbol;

[0100] The fifth symbol;

[0101] The sixth symbol is either the first or last symbol received in the current CSI.

[0102] The seventh symbol is either the first or last symbol corresponding to the CSI transmission period indicated by the CSI configuration information;

[0103] The eighth symbol is either the first or last symbol of the deactivation signaling corresponding to the CSI configuration information.

[0104] In some embodiments, if the CPU and / or the target resources are insufficient, the target CSI configuration information is determined according to the priority order of the CPU and / or the target resources and the CSI configuration information.

[0105] In some embodiments, the priority order of the CSI configuration information is determined according to at least one of the following:

[0106] The periodic information in the CSI configuration information;

[0107] The CSI configuration information specifies the triggering method for sending CSI.

[0108] The feedback quantity in the CSI configuration information;

[0109] Information in the CSI configuration information;

[0110] The number of reference signals corresponding to the CSI configuration information;

[0111] High-level parameter information associated with the CSI configuration information;

[0112] Serving the community;

[0113] The identifier of the CSI configuration information.

[0114] Thirdly, embodiments of this disclosure provide a channel state information processing apparatus applied to a terminal device. The channel state information processing apparatus includes a receiving module and a transmitting module, wherein:

[0115] The receiving module is used to receive first information sent by the network device, the first information being used to trigger and / or activate Channel State Information (CSI) configuration information;

[0116] The sending module is used to send the CSI to the network device. The CSI is sent based on the CSI processing unit CPU, the target resource, and the CSI configuration information. The target resource is associated with the AI ​​processing of the CSI.

[0117] Fourthly, embodiments of this disclosure provide a channel state information processing apparatus applied to a network device. The channel state information processing apparatus includes a transmitting module and a receiving module, wherein:

[0118] The sending module is used to send first information to the terminal device, the first information being used to trigger and / or activate Channel State Information (CSI) configuration information;

[0119] The receiving module is used to receive CSI sent by the terminal device. The CSI is sent based on the CPU, target resources, and CSI configuration information. The target resources are associated with the AI ​​processing of the CSI.

[0120] Fifthly, embodiments of this disclosure provide a terminal device, including a memory, a transceiver, and a processor:

[0121] The memory is used to store computer programs;

[0122] The transceiver is used to send and receive data under the control of the processor;

[0123] The processor is configured to read the computer program from the memory and perform the following operations:

[0124] Receive first information sent by the network device, the first information being used to trigger and / or activate Channel State Information (CSI) configuration information;

[0125] The CSI is sent to the network device. The CSI is sent based on the CSI processing unit CPU, the target resource, and the CSI configuration information. The target resource is associated with the AI ​​processing of the CSI.

[0126] Sixthly, embodiments of this disclosure provide a network device, including a memory, a transceiver, and a processor:

[0127] The memory is used to store computer programs;

[0128] The transceiver is used to send and receive data under the control of the processor;

[0129] The processor is configured to read the computer program from the memory and perform the following operations:

[0130] The first information sent to the terminal device is used to trigger and / or activate the Channel State Information (CSI) configuration information.

[0131] The terminal device receives a CSI sent by the terminal device. The CSI is sent based on the CPU, target resources, and CSI configuration information. The target resources are associated with the AI ​​processing of the CSI.

[0132] In a seventh aspect, this disclosure provides a processor-readable storage medium storing a computer program configured to cause a processor to perform the method described in the first aspect or to perform the method described in the second aspect.

[0133] This disclosure provides a channel state information (CSI) processing method, apparatus, device, and storage medium. A terminal device can receive first information sent by a network device, wherein the first information can be used to trigger and / or activate CSI configuration information. The terminal device can send CSI to the network device, wherein the CSI is sent based on CPU, target resources, and CSI configuration information, and the target resources are associated with the AI ​​processing of the CSI. In the above method, the terminal device can accurately determine the capability requirements for CSI feedback based on the target resources, which helps in the allocation of communication resources. Furthermore, the terminal device can comprehensively and accurately assess the resources required to calculate the CSI based on CPU and target resources, thus avoiding CSI calculation failures, improving the success rate of CSI calculation, and improving the success rate and accuracy of CSI transmission.

[0134] It should be understood that the description in the foregoing summary section is not intended to limit the key or essential features of the embodiments of this disclosure, nor is it intended to restrict the scope of this disclosure. Other features of this disclosure will become readily apparent from the following description. Attached Figure Description

[0135] To more clearly illustrate the technical solutions in this disclosure or related technologies, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of this disclosure. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0136] Figure 1 is a schematic diagram of a communication system provided in an embodiment of this disclosure;

[0137] Figure 2 is a schematic diagram of a channel state information processing method provided in an embodiment of this disclosure;

[0138] Figure 3 is a schematic diagram of another channel state information processing method provided in an embodiment of this disclosure;

[0139] Figure 4 is a schematic diagram of the maximum CPU required for CSI reporting according to an embodiment of this disclosure;

[0140] Figure 5 is a schematic diagram of the time occupied according to an embodiment of this disclosure;

[0141] Figure 6 is a schematic diagram of a method for determining target CSI configuration information provided in an embodiment of this disclosure;

[0142] Figure 7 is a schematic diagram of a method for switching CSI transmission provided in an embodiment of this disclosure;

[0143] Figure 8 is a schematic diagram of a method for sending second information according to an embodiment of this disclosure;

[0144] Figure 9 is a schematic diagram of the structure of a channel state information processing device provided in an embodiment of this disclosure;

[0145] Figure 10 is a schematic diagram of another channel state information processing device provided in an embodiment of this disclosure;

[0146] Figure 11 is a schematic diagram of the structure of a terminal device provided in an embodiment of this disclosure;

[0147] Figure 12 is a schematic diagram of the structure of a network device provided in an embodiment of this disclosure. Detailed Implementation

[0148] In this disclosure, the term "and / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent three cases: A alone, A and B simultaneously, and B alone. The character " / " generally indicates that the preceding and following related objects have an "or" relationship.

[0149] In this disclosure, the term "multiple" refers to two or more, and other quantifiers are similar.

[0150] The technical solutions of the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this disclosure, and not all embodiments. Based on the embodiments of this disclosure, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this disclosure.

[0151] This disclosure provides a channel state information processing method, apparatus, device, and storage medium. The CSI can be transmitted based on CPU, target resources, and CSI configuration information. Furthermore, the target resources can be associated with the AI ​​processing of the CSI. Therefore, the terminal device can comprehensively and accurately assess the resources required to calculate the CSI based on the CPU and target resources, thereby improving the success rate of CSI calculation and transmission.

[0152] The method and apparatus are based on the same technical concept. Since the methods and apparatus solve problems in similar ways, the implementation of the apparatus and methods can refer to each other, and the repeated parts will not be described again.

[0153] The technical solutions provided in this disclosure can be applied to a variety of systems. For example, applicable systems may include Long Term Evolution (LTE) systems, LTE Frequency Division Duplex (FDD) systems, LTE Time Division Duplex (TDD) systems, Long Term Evolution Advanced (LTE-A) systems, Universal Mobile Telecommunications System (UMTS), Worldwide Interoperability for Microwave Access (WiMAX) systems, 5G New Radio (NR) systems and their evolved communication systems, and 6G (sixth generation mobile communication technology) systems. These systems may include terminal equipment and network equipment. The systems may also include a core network component, such as the Evolved Packet Core (EPC) and the 5G Core Network (5GC).

[0154] The terminal devices involved in the embodiments of this disclosure can be devices that provide voice and / or data connectivity to users, handheld devices with wireless connectivity, or other processing devices connected to a wireless modem. The names of the terminal devices may differ in different systems; for example, in 5G or 6G systems, the terminal device may be called User Equipment (UE). Wireless terminal devices can be USB storage devices, other personal computer memory devices, and dongles. They can also communicate with one or more core networks (CNs) via a Radio Access Network (RAN). Wireless terminal devices can be mobile terminal devices, such as mobile phones (or "cellular" phones) and computers with mobile terminal devices. For example, they can be portable, pocket-sized, handheld, computer-embedded, or vehicle-mounted mobile devices that exchange voice and / or data with the radio access network. Examples of such devices include Personal Communication Service (PCS) phones, cordless phones, Session Initiated Protocol (SIP) phones, Wireless Local Loop (WLL) stations, Personal Digital Assistants (PDAs), personal computers, tablets, and Machine-type Communication (MTC) terminal devices. Wireless terminal devices can also be referred to as systems, subscriber units, subscriber stations, mobile stations, mobile devices, remote stations, access points, remote terminals, access terminals, user terminals, user agents, user devices, and wireless access devices and routers / modems that meet the limitations of this definition, but are not limited to these specific embodiments in this disclosure.

[0155] The network device involved in this disclosure can be a base station, which may include multiple cells providing services to terminals. Depending on the application, the base station may also be called an access point, or a device in the access network that communicates with wireless terminal devices through one or more sectors on the air interface, or other names. The network device can be used to exchange received air frames with Internet Protocol (IP) packets, acting as a router between the wireless terminal device and the rest of the access network, where the rest of the access network may include an Internet Protocol (IP) communication network. The network device can also coordinate the attribute management of the air interface. For example, the network device involved in this disclosure can be an evolved Node B (eNB or e-NodeB) in a long term evolution (LTE) system, a 5G base station (gNB) in a next generation system, or a Home evolved Node B (HeNB), relay node, femto, pico, network testing equipment, etc., and is not limited in this disclosure. In some network architectures, network devices may include centralized unit (CU) nodes and distributed unit (DU) nodes, which may also be geographically separated.

[0156] The communication system of this disclosure embodiment will now be described with reference to FIG1.

[0157] Figure 1 is a schematic diagram of a communication system provided in an embodiment of this disclosure. Referring to Figure 1, the system includes a network device and a terminal device. The network device and the terminal device can pre-agree on multiple CSI configuration information, or the network device can pre-send multiple CSI configuration information to the terminal device. The network device can send an instruction to the terminal device to activate the CSI configuration information. The terminal device can determine the activated CSI configuration information based on this instruction. The terminal device can calculate the CSI based on the activated CSI configuration information and send the CSI to the network device.

[0158] In related technologies, network devices can activate and / or trigger multiple CSI configuration information in terminal devices. Therefore, when multiple CSI configuration information in a terminal device is activated and / or triggered, the terminal device needs to determine whether its processing capacity is sufficient to send all CSIs. In some implementations, the terminal device can calculate CSIs based on CPU. Therefore, when the calculation times of multiple CSIs overlap, the CPU required for the terminal device to calculate multiple CSIs cannot exceed the CPU limit of the terminal device. For example, if a network device activates 10 CSI configuration information, and the calculation times of 7 CSI configuration information overlap, the terminal device can determine whether the CPU required to calculate these 7 CSIs exceeds the CPU limit. If it does not exceed the limit, the terminal device can calculate all CSIs. However, future communication systems may also include AI-based CSIs. The capability requirements for calculating this type of CSI differ from those for calculating traditional CSIs. Since CPU generally represents the capability requirements for calculating traditional CSIs, it cannot represent the capability requirements for calculating AI-based CSIs. Therefore, the accuracy of the terminal device's assessment of the resources required to calculate CSIs is low, which may lead to CSI calculation failures and a low success rate for CSI transmission.

[0159] To address the aforementioned technical problems, this disclosure provides a Channel State Information (CSI) processing method. A terminal device can receive first information from a network device for activating and / or triggering CSI configuration information, and then send CSI to the network device. The CSI is sent based on CPU, target resources, and the CSI configuration information. In this method, the target resources may include at least one of the following: computing power resources for AI data processing, memory resources for AI data processing, cache resources for AI data processing, and storage space resources for AI data processing. This allows the terminal device to comprehensively and accurately assess the resources required to calculate the CSI based on CPU and target resources, thus avoiding CSI calculation failures, improving the success rate of CSI calculation, and enhancing the success rate and accuracy of CSI transmission.

[0160] The channel state information processing method provided in this disclosure will now be described with reference to embodiments. It is understood that this disclosure uses network devices and terminal devices as examples to illustrate the interaction, but this disclosure does not limit the execution subject of the interaction. For example, the method executed by the network device in the embodiments of this disclosure can also be implemented by modules (e.g., circuits, chips, or chip systems) in the network device. Similarly, the method executed by the terminal device in the embodiments of this disclosure can also be implemented by modules (e.g., circuits, chips, or chip systems) in the terminal device. The following description uses terminal devices and network devices as examples.

[0161] Figure 2 is a schematic diagram of a channel state information processing method provided in an embodiment of this disclosure. Referring to Figure 2, the channel state information processing method includes:

[0162] S201, The terminal device receives the first information sent by the network device.

[0163] The first information can be used to trigger and / or activate CSI configuration information.

[0164] In some embodiments, CSI configuration information can be used to calculate and transmit CSI. For example, CSI configuration information may include CSI resource configuration, CSI reporting configuration, etc., wherein CSI resource configuration can be used to configure the reference signal for calculating CSI, and CSI resource configuration may include the ID of the bandwidth portion, a resource list for beam management and reported CSI interference measurements, etc. CSI reporting configuration can be used to configure the behavior of reporting CSI, and CSI reporting configuration may include the ID of CSI reporting configuration, resources for channel measurement, etc.

[0165] In some embodiments, the first information may be triggering signaling and / or activating signaling, wherein the triggering signaling and / or activating signaling may include an identifier of CSI configuration information, which may indicate the CSI configuration information that is triggered and / or activated.

[0166] For example, the CSI configuration information in the terminal device may include CSI configuration information 1, CSI configuration information 2, and CSI configuration information 3. The network device may send an activation signaling to the terminal device. If the activation signaling includes the identifier of CSI configuration information 1, the terminal device may activate CSI configuration information 1 after receiving the activation signaling. If the activation signaling includes the identifiers of CSI configuration information 2 and CSI configuration information 3, the terminal device may activate CSI configuration information 2 and CSI configuration information 3 after receiving the activation signaling.

[0167] In some embodiments, the first information may also be indication information, which may indicate CSI configuration information. The terminal device may determine the CSI configuration information indicated by the indication information as the CSI configuration information to be triggered and / or activated. Optionally, the first information may also be any other information used to activate and / or trigger CSI configuration information, which is not limited in this embodiment.

[0168] Optionally, the network device may send the first information based on Radio Resource Control (RRC) signaling, or it may send the first information based on any other feasible implementation method. This embodiment of the present disclosure does not limit this.

[0169] S202, The terminal device sends a CSI to the network device.

[0170] CSI is sent based on CPU, target resources, and CSI configuration information.

[0171] The target resource is associated with the AI ​​processing of CSI. For example, the target resource can be resources and / or parameters related to the AI ​​processing of CSI. For instance, the processing power requirements of the terminal device for an AI model differ from those of traditional CSI feedback. The processing power requirements of the terminal device for an AI model can also include memory requirements, computing power requirements, cache requirements, and storage space requirements. The resources required for such requirements can be the target resource or the AI ​​CPU.

[0172] In some embodiments, CSI configuration information may indicate the method of calculating CSI and the method of sending CSI. When the terminal device calculates CSI, the terminal device needs to use CPU and target resources. Furthermore, the CPU used by the terminal device to calculate CSI needs to be less than the CPU limit of the terminal device, and the target resources used by the terminal device to calculate CSI need to be less than the target resource limit of the terminal device.

[0173] For example, the terminal device may include 10 CPUs and 10 AI CPUs (for example only), and the CSI configuration information activated by the network device includes CSI configuration information 1 and CSI configuration information 2. If the terminal device requires 3 CPUs and 3 AI CPUs to report the CSI corresponding to CSI configuration information 1, and requires 5 CPUs and 5 AI CPUs to report the CSI corresponding to CSI configuration information 2, the terminal device can report the CSI corresponding to both CSI configuration information 1 and CSI configuration information 2. If the terminal device requires 7 CPUs and 5 AI CPUs to report CSI configuration information 1, and 5 CPUs and 3 AI CPUs to report CSI configuration information 2, or if the terminal device requires 3 CPUs and 7 AI CPUs to report CSI configuration information 1, and 5 CPUs and 5 AI CPUs to report CSI configuration information 2, then the terminal device's CPUs and AI CPUs cannot support reporting 2 CSIs. Therefore, the terminal device cannot report both CSI configuration information 1 and CSI configuration information 2 simultaneously. The terminal device can report either CSI configuration information 1 or CSI configuration information 2.

[0174] This disclosure provides a channel state information (CSI) processing method. A terminal device can receive first information sent by a network device and send CSI to the network device. In this method, CSI is sent based on CPU, target resources, and CSI configuration information. The target resources are associated with the AI ​​processing of CSI. In this way, the terminal device can accurately determine the capability requirements for CSI feedback based on the target resources, which helps in the allocation of communication resources. Furthermore, the terminal device can comprehensively and accurately evaluate the resources required to calculate CSI based on CPU and target resources. This can avoid CSI calculation failures, prioritize the calculation of more important CSI information, improve the success rate of CSI calculation, and improve the success rate and accuracy of CSI transmission.

[0175] Based on the embodiment shown in Figure 2, another channel state information processing method will be described below with reference to Figure 3.

[0176] Figure 3 is a schematic diagram of another channel state information processing method provided in an embodiment of this disclosure. Referring to Figure 3, the method flow includes:

[0177] S301. The terminal device sends the upper limit of the CPU and / or the upper limit of the target resources to the network device.

[0178] The CPU limit can be the maximum number of CPUs available for the terminal device to process CSI-related tasks (such as calculating CSI, reporting CSI, etc.). For example, if the terminal device includes 10 CPUs, and these 10 CPUs can be used to process CSI-related tasks, then the CPU limit for the terminal device can be 10. Alternatively, if the terminal device includes 10 CPUs, and 8 of them can be used to process CSI-related tasks, then the CPU limit for the terminal device can be 8.

[0179] The upper limit of the target resources can be the maximum target resources available to the terminal device when processing CSI-related tasks. Target resources may include at least one of the following:

[0180] Computing resources used for AI data processing;

[0181] Memory resources used for AI data processing;

[0182] Cache resources used for AI data processing;

[0183] Storage space resources used for AI data processing.

[0184] In some embodiments, the upper limit of the target resource may include at least one of the following:

[0185] The maximum computing power resources used for AI data processing;

[0186] Maximum memory resources used for AI data processing;

[0187] Maximum cache resources used for AI data processing;

[0188] Maximum storage space resources for AI data processing.

[0189] Computing resources can refer to the available computing power for AI data processing. For example, computing resources can be represented by the computational complexity of AI data processing. For instance, the higher the complexity of the AI ​​model when processing data, the greater the computing resources required. For example, computing resources can be 10 trillion floating-point operations per second (TFLOPS) or 30 TFLOPS. For example, if the maximum computing power of a terminal device for AI data processing is 30 TFLOPS, then the computing resources in the target resource specifically for AI data processing are also 30 TFLOPS.

[0190] In some embodiments, the maximum sum of floating-point operations (FLOPs) of all AI models running on the terminal device within the same time slot can be considered as the computing power resource of the terminal device used for AI data processing. For example, if the FLOPs of all AI models running on the terminal device in time slot 1 are 10 trillion floating-point operations per second, and the FLOPs of all AI models running in time slot 2 are 30 trillion floating-point operations per second, then the computing power resource of the terminal device used for AI data processing is 30 trillion floating-point operations per second.

[0191] Memory resources can be defined as the available memory used for AI data processing. For example, AI models consume memory when performing inference and / or prediction; therefore, memory resources can be defined as the maximum memory available for AI model inference and / or prediction. For instance, if the maximum memory allowed for multiple AI models to use is 8GB when multiple AI models are performing inference and / or prediction, then the memory resources in the target resource allocated to AI data processing are 8GB.

[0192] The cache resource can be the available cache for AI data processing. For example, an AI model will occupy the cache when making predictions (e.g., an AI model can make CSI predictions based on cached data). Therefore, the cache resource can be the maximum cache available for AI model predictions. For example, if the maximum cache allowed for multiple AI models to occupy by the terminal device is 5MB when multiple AI models are performing inference and / or prediction, then the cache resource in the target resource used for AI data processing is 5MB.

[0193] Storage space resources can refer to the available storage space used for AI data processing. For example, before an AI model is used, the terminal device can load the structure and parameters of the AI ​​model. The parameters of the AI ​​model can be stored in the storage space; therefore, the storage space resources can be the maximum storage space used by the AI ​​model. For example, if the terminal device has 100GB of storage space for storing the AI ​​model, then the storage space resources used for AI data processing in the target resource are 100GB. For example, if the terminal device is allowed to store 100GB of AI model data in the same time slot, then the storage space resources used for AI data processing in the target resource are 100GB.

[0194] In some embodiments, the type of AI data processing includes at least one of the following:

[0195] Perform inference and / or prediction based on AI data using AI models;

[0196] AI data is used to perform AI model training tasks or related data collection.

[0197] Monitor the performance of AI model inference and / or prediction.

[0198] In some embodiments, after the terminal device collects data, it can input the data into the AI ​​model, which can then perform reasoning and / or prediction based on the data. Therefore, AI data processing can be used to perform reasoning or prediction based on AI data.

[0199] In some embodiments, the terminal device can collect data for AI model training, and after collecting the data, the terminal device can also train (or train with the help of other entities) the AI ​​model. Therefore, AI data processing can be for performing AI model training tasks based on AI data, or for performing data collection related to AI model training tasks.

[0200] In some embodiments, the performance of an AI model may change during inference and / or prediction, requiring timely stopping, updating, or switching of the AI ​​model. Therefore, AI data processing can be used to monitor the performance of AI model inference and / or prediction.

[0201] S302, The network device sends the first information to the terminal device.

[0202] The first information is used to trigger and / or activate the CSI configuration information in the terminal device.

[0203] S303, The terminal device sends a CSI to the network device.

[0204] The terminal device can send CSI to the network device in the following feasible implementation: determine the target CSI configuration information in the triggered and / or activated CSI configuration information, and send CSI to the network device according to the target CSI configuration information.

[0205] In some embodiments, the network device may trigger and / or activate multiple CSI configuration information in the terminal device. However, due to the limited CPU and target resources in the terminal device, the terminal device may not be able to report all CSI configuration information corresponding to the CSI. Therefore, the terminal device may determine the target CSI configuration information, wherein the CSI corresponding to the target CSI configuration information is the CSI that the terminal device can report.

[0206] For example, when the terminal device's CPU and target resources support reporting CSIs corresponding to all activated CSI configuration information, the terminal device can determine all activated or triggered CSI configuration information as the target CSI configuration information, and calculate and report CSIs based on the target CSI configuration information. When the terminal device's CPU and target resources cannot support reporting CSIs corresponding to all activated CSI configuration information, the terminal device can determine the target CSI configuration information from all activated or triggered CSI configuration information, and calculate and report CSIs based on the target CSI configuration information.

[0207] In some embodiments, the target CSI configuration information is determined based on the target resources and / or CPU occupied by each CSI configuration information, and the target time for occupying the target resources and / or CPU.

[0208] In some embodiments, the terminal device can determine the target resources and / or CPU occupied by each CSI configuration information, and determine the overlap time of target resources and / or CPU occupied by multiple CSI configuration information according to the target time of occupying target resources and / or CPU. The maximum target resources and / or CPU corresponding to the overlap time is the target resources and CPU required by the terminal device to report all CSIs. The terminal device can determine the target CSI configuration information according to the target resources and CPU required to report all CSIs.

[0209] Below, referring to Figure 4, we will explain the CPU required for terminal devices to report CSI, taking the CPU as an example.

[0210] Figure 4 is a schematic diagram illustrating the maximum CPU required for CSI reporting according to an embodiment of this disclosure. Referring to Figure 4, it includes: a timeline, CSI configuration information 1, CSI configuration information 2, and CSI configuration information 3. Specifically, CSI configuration information 1, when activated, occupies 3 CPUs; CSI configuration information 2, when activated, occupies 5 CPUs; and CSI configuration information 3, when activated, occupies 7 CPUs.

[0211] Please refer to Figure 4. There is an overlap in the usage time between CSI configuration information 1 and CSI configuration information 2. During this overlap, CSI configuration information 1 and CSI configuration information 2 will simultaneously occupy 8 CPUs. There is also an overlap in the usage time between CSI configuration information 2 and CSI configuration information 3. During this overlap, CSI configuration information 2 and CSI configuration information 3 will simultaneously occupy 12 CPUs.

[0212] Please refer to Figure 4. The terminal device (not shown in Figure 4) requires 12 CPUs to report the three CSIs corresponding to CSI configuration information 1, CSI configuration information 2 and CSI configuration information 3. That is, when the terminal device has 12 CPUs, the terminal device can report 3 CSIs. When the number of CPUs in the terminal device is less than 12, the terminal device cannot report 3 CSIs.

[0213] In the embodiment shown in Figure 4, the CPU can also be a computing power resource, memory resource, cache resource, and storage space resource used for AI data processing. The method by which the terminal device determines the maximum computing power resource, memory resource, cache resource, and storage space resource required for reporting CSI is the same as the method for determining the maximum CPU, and will not be described again in this embodiment.

[0214] In some embodiments, CSI reporting may involve inference and / or prediction of multiple AI models. Therefore, the terminal device determines that the target resources and CPUs used for CSI reporting can be the sum of the target resources and CPUs used for inference and / or prediction of multiple AI models. For example, if CSI reporting requires AI model 1 and AI model 2, and if AI model 1 requires 3 CPUs and 5 AI CPUs for inference, and AI model 2 requires 4 CPUs and 3 AI CPUs for inference, then the terminal device can determine that the CSI reporting requires 7 CPUs and 8 AI CPUs. The AI ​​CPUs can include computing power resources, memory resources, cache resources, and storage space resources used for AI data processing.

[0215] In some embodiments, since memory and cache may change during AI model inference, the memory occupied by the AI ​​model inference can be the maximum memory occupied during the AI ​​model inference process, and the cache occupied by the AI ​​model inference can be the maximum cache occupied during the AI ​​model inference process. For example, the AI ​​model inference may initially occupy 2GB of memory and 1MB of cache, and during the AI ​​model inference process, it may occupy 8GB of memory and 10MB of cache. The terminal device can determine that the AI ​​model inference occupies 8GB of memory and 10MB of cache.

[0216] In some embodiments, the target resources and / or CPU occupied by the CSI configuration information are determined based on at least one of the following:

[0217] Feedback quantity in CSI configuration information;

[0218] High-level parameters associated with CSI configuration information;

[0219] Parameters in CSI configuration information;

[0220] Reference signal resources corresponding to CSI configuration information;

[0221] Indicators or associated high-level parameters used to indicate the capabilities of terminal devices.

[0222] In some embodiments, the feedback quantities in the CSI configuration information may include a Channel Quality Indicator (CQI), a Precoding Matrix Indicator (PMI), and a Rank Indicator (RI), which are not limited in this disclosure.

[0223] In some embodiments, the high-level parameters associated with CSI configuration information may include the subband for calculating PMI, the time-domain type of CSI reporting (e.g., periodic reporting, aperiodic reporting, and semi-continuous reporting), the method for triggering CSI reporting (e.g., whether CSI reporting is triggered based on events), and high-level parameters such as the model (structure) and dataset ID related to AI use cases. This disclosure does not limit these parameters.

[0224] In some embodiments, parameters in the CSI configuration information may include codebook type or other parameters that can be used to indicate the CSI configuration information. For example, parameters that can be used to indicate the CSI configuration information may include those indicating the time relationship corresponding to the reported content of the CSI prediction (the size of the prediction time window, the predicted future time point), and IDs associated with AI model beam management (e.g., associated AI model ID, dataset ID, etc.).

[0225] In some embodiments, the reference signal resources corresponding to the CSI configuration information may include the type of reference signal, the number of resources (e.g., the number of reference signal resources used for CSI measurement and interference measurement), periodicity (e.g., periodic, aperiodic, or semi-continuous), etc., wherein the type of reference signal may include the reference signal for channel state information measurement (CSI-RS), the reference signal for interference measurement (CSI-IM), etc.

[0226] In some embodiments, metrics or associated high-level parameters indicating the capabilities of a terminal device may be sent by the terminal device. These metrics or parameters may be associated with computing power resources, memory resources, cache resources, and storage space resources for AI data processing. For example, the terminal device may determine a metric or parameter based on the computing power resources, memory resources, cache resources, and storage space resources for AI data processing, which may indicate the capabilities of the terminal device.

[0227] In some embodiments, the terminal device may process at least one of the following based on a pre-trained model: feedback quantity in CSI configuration information, high-level parameters associated with CSI configuration information, parameters in CSI configuration information, reference signal resources corresponding to CSI configuration information, and indicators (or associated high-level parameters) used to indicate the capabilities of the terminal device, to obtain the target resources and / or CPU occupied by CSI configuration information.

[0228] In some embodiments, the terminal device may also process at least one of the following based on any feasible implementation method: feedback quantity in CSI configuration information, higher-level parameters associated with CSI configuration information, parameters in CSI configuration information, reference signal resources corresponding to CSI configuration information, and indicators (or associated higher-level parameters) used to indicate the capabilities of the terminal device, to obtain the target resources and / or CPU occupied by CSI configuration information. This disclosure does not limit this aspect.

[0229] The target time can be the time that the target resources and / or CPU are occupied, corresponding to the CSI configuration information.

[0230] In some embodiments, since the terminal device can determine the target resources and CPU required to report CSI based on the target resources and CPU in the overlapping occupancy time, the terminal device can determine the target time corresponding to each CSI configuration information, and thus determine the overlapping occupancy time.

[0231] The target time may include a start time and an end time. In some embodiments, the start time is determined based on at least one of the following:

[0232] First symbol;

[0233] Second symbol;

[0234] Third symbol;

[0235] Fourth symbol;

[0236] The fifth symbol.

[0237] In some embodiments, the first symbol can be the first symbol that triggers and / or activates the CSI configuration information, the last symbol, or the symbol following the last symbol. For example, after the CSI configuration information is activated, it will occupy CPU and / or target resources. Therefore, the start time of the target time can be the first symbol, the last symbol, or the symbol following the last symbol after the CSI configuration information is activated. In these three symbols, target resources and / or CPU may be occupied.

[0238] In some embodiments, the second symbol is the first symbol, the last symbol, or the symbol following the last symbol of the transmission timing of the first reference signal corresponding to the current CSI. For example, after receiving the first reference signal, the terminal device can perform model inference or performance monitoring corresponding to the current CSI. Therefore, the CSI configuration information can occupy CPU and / or target resources. The start time of the target time can be the first symbol, the last symbol, or the symbol following the last symbol of the transmission timing of the first reference signal. In these three symbols, target resources and / or CPU may be occupied.

[0239] In some embodiments, the third symbol can be the first symbol, the last symbol, or the symbol following the last symbol of the transmission of the last reference signal corresponding to the current CSI. For example, after receiving the last reference signal, the terminal device can perform model inference or performance monitoring corresponding to the current CSI. Therefore, the CSI configuration information can occupy CPU and / or target resources. The start time of the target time can be the first symbol, the last symbol, or the symbol following the last symbol of the transmission of the last reference signal. In these three symbols, target resources and / or CPU may be occupied.

[0240] In some embodiments, the fourth symbol can be the first symbol after the measurement at the last reference signal transmission time is completed, or the first symbol to begin processing the measurement result obtained after the measurement. For example, if the terminal device determines the CSI based on a non-AI prediction method, the CPU will be occupied during the measurement at the reference signal transmission time. After the measurement at the last reference signal transmission time is completed, the terminal device can compress the CSI based on an AI model. Therefore, the first symbol after the measurement can be the start time of occupying the target resource or the end time of occupying the CPU. Similarly, the first symbol to begin processing the measurement result obtained after the measurement can also be the start time of occupying the target resource or the end time of occupying the CPU.

[0241] In some embodiments, the fifth symbol can be the first symbol after completing AI model inference and / or prediction, or the Mth symbol after completing the measurement of the reference signal transmission timing. For example, M is associated with the terminal device capability; the higher the terminal device capability, the smaller M is, and the lower the terminal device capability, the larger M is. For example, the terminal device can predict channel quality based on AI and calculate PMI, CQI, and RI based on a non-AI codebook. This process consumes target resources during AI-based prediction and CPU resources during the calculation of PMI, CQI, and RI. Therefore, the first symbol after the terminal device completes AI model inference or the measurement of the reference signal transmission timing can be the end time of target resource consumption or the start time of CPU consumption. Alternatively, the terminal device can start calculating PMI, CQI, and RI at the Mth symbol after completing AI model inference or the measurement of the reference signal transmission timing. Therefore, the Mth symbol can be the end time of target resource consumption or the start time of CPU consumption.

[0242] In some embodiments, the end time is determined based on at least one of the following:

[0243] Fourth symbol;

[0244] The fifth symbol;

[0245] The sixth symbol;

[0246] The seventh symbol;

[0247] The eighth symbol.

[0248] In some embodiments, the fourth and fifth symbols may also represent the end time of occupying target resources and / or CPU.

[0249] In some embodiments, the sixth symbol can be either the first or the last symbol of the current CSI transmission. For example, when the terminal device transmits the CSI, the CSI calculation has already been completed; therefore, the first symbol can be the end time of the terminal device's occupation of target resources and / or CPU. For example, the terminal device releases the occupied memory resources at the end of the CSI transmission; therefore, the last symbol can be the end time of the terminal device's occupation of target resources and / or CPU.

[0250] In some embodiments, the seventh symbol is the first or last symbol corresponding to the transmission period of the CSI indicated by the CSI configuration information. For example, when the CSI configuration information indicates that the CSI is transmitted non-periodically, the CSI configuration information may include the transmission time of the CSI, and the first or last symbol of the transmission time may be the end time of the terminal device occupying the target resources and / or CPU.

[0251] In some embodiments, the eighth symbol can be the first or last symbol of the deactivation signaling corresponding to the CSI configuration information. For example, the deactivation signaling can be used to deactivate the CSI configuration information. After the CSI configuration information is deactivated, it will not occupy the target resources and / or CPU. Therefore, the first symbol of the deactivation signaling can be the end time of the terminal device occupying the target resources and / or CPU, and the last symbol of the deactivation signaling can also be the end time of the terminal device occupying the target resources and / or CPU.

[0252] In some embodiments, since the target resources may include computing power resources for AI data processing, memory resources for AI data processing, cache resources for AI data processing, and storage space resources for AI data processing, the time occupied by computing power resources, memory resources, cache resources, and space resources during AI model inference and / or prediction is also different. The time occupied by computing power resources, memory resources, cache resources, and space resources will be explained below with reference to Figure 5.

[0253] It should be noted that the start time and end time can also be other time units, such as seconds, milliseconds, microseconds, or time measurement units of communication systems such as time slots, symbols, frames, subframes, miniframes, etc., or absolute values. This disclosure does not limit these.

[0254] Figure 5 is a schematic diagram illustrating the time occupancy provided in an embodiment of this disclosure. Referring to Figure 5, it includes a timeline. The timeline includes transmission timing 1, transmission timing 2, transmission timing 3, transmission timing 4, transmission timing 5, and the time for sending the CSI. During the period between transmission timing 5 and sending the CSI, the terminal device (not shown in Figure 5) processes the CSI task, occupying AICPU1 and AICPU2. During the period between transmission timing 1 and sending the CSI, the terminal device processes the CSI task, occupying AICPU3. During the period between transmission timing 1 and the completion of sending the CSI, the terminal device processes the CSI task, occupying AICPU4.

[0255] In the embodiment shown in Figure 5, AICPU1 represents the computing power resources (complexity) required for CSI transmission, AICPU2 represents the memory resources (e.g., running memory) required for CSI transmission, AICPU3 represents the cache resources (e.g., cache queue space) required for AI processing, and AICPU4 represents the storage space occupied by AI model loading.

[0256] In the embodiment shown in Figure 5, the occupancy of AICPU1 and AICPU2 may begin from the first symbol after the last symbol of the last reference signal transmission timing corresponding to the CSI transmission and continue until the first symbol of the current CSI transmission. The occupancy of AICPU3 may begin from the first symbol of the first reference signal transmission timing used to calculate the CSI transmission and continue until the first symbol of the current CSI transmission. The occupancy of AICPU4 may begin from the first symbol after the last symbol of triggering and / or activating the CSI configuration information and continue until the last symbol of deactivating the current CSI configuration information.

[0257] This disclosure provides a channel state information (CSI) processing method. A terminal device sends its CPU upper limit and / or target resource upper limit to a network device. The network device sends first information to the terminal device. The terminal device determines the target CSI configuration information from the triggered and / or activated CSI configuration information and sends the CSI to the network device based on the target CSI configuration information. In this method, the terminal device, based on CPU and target resources, can comprehensively and accurately assess the resources required to calculate the CSI. This avoids CSI calculation failures, improves the success rate of CSI calculation, and enhances the success rate and accuracy of CSI transmission.

[0258] Based on any of the above embodiments, when the CPU and / or target resources are insufficient, the terminal device can determine the target CSI configuration information according to the priority order of the CPU and / or target resources and the CSI configuration information. The method for the terminal device to determine the target CSI configuration information will be described below with reference to Figure 6.

[0259] Figure 6 is a schematic diagram of a method for determining target CSI configuration information provided in an embodiment of this disclosure. Referring to Figure 6, it includes:

[0260] S601. Determine the CSI configuration information set according to priority order.

[0261] Priority order can indicate the priority of multiple CSI configuration information. In some embodiments, CSI tasks corresponding to high-priority CSI configuration information can be processed first, while CSI tasks corresponding to low-priority CSI configuration information can be processed later.

[0262] In some embodiments, the priority order can be pre-agreed upon.

[0263] In some embodiments, the priority order of CSI configuration information is determined according to at least one of the following:

[0264] Periodic information in CSI configuration information;

[0265] The CSI configuration information specifies the triggering method for sending CSI messages;

[0266] Feedback quantity in CSI configuration information;

[0267] Information in CSI configuration information;

[0268] The number of reference signals corresponding to the CSI configuration information;

[0269] High-level parameter information associated with CSI configuration information;

[0270] Serving the community;

[0271] CSI configuration information identifier.

[0272] Periodic information can include periodic CSI transmission and non-periodic CSI transmission. When the terminal device determines the priority of CSI configuration information based on the periodic information of the CSI configuration information, non-periodic CSI transmission takes precedence over semi-persistent CSI transmission, and semi-persistent CSI transmission takes precedence over periodic CSI transmission.

[0273] For example, the periodic information in CSI configuration information 1 is sent aperiodically, the periodic information in CSI configuration information 2 is sent periodically, and the periodic information in CSI configuration information 3 is sent semi-continuously. The terminal device can determine that the priority of CSI configuration information 1 is higher than the priority of CSI configuration information 3, and the priority of CSI configuration information 3 is higher than the priority of CSI configuration information 2.

[0274] The CSI configuration information allows for either event-triggered or non-event-triggered CSI sending. For example, event-triggered CSI can be sent after the terminal device determines that a preset event has occurred, while non-event-triggered CSI can be sent based on the CSI configuration information.

[0275] In some embodiments, event-triggered CSI transmission takes precedence over non-periodic CSI transmission.

[0276] In some embodiments, the feedback quantities in the CSI configuration information may include CQI, PMI, RI, performance monitoring results, other parameters of the CSI configuration information, other high-level parameters, or other predefined values.

[0277] For example, when the terminal device determines the priority of CSI configuration information based on the feedback quantity in the CSI configuration information, the feedback quantity of the CSI configuration information includes L1-Signal to Interference plus Noise Ratio (SINR) or L1-Reference Signal Received Power (RSRP), and the CSI configuration information with priority is given priority; while when the feedback quantity is set to None, it means that the CSI configuration information is used for data collection and has the lowest priority.

[0278] In some embodiments, the information in the CSI configuration information may also include sub-use cases identified by codebookType, reference signal resource sets, and IDs of other associated CSI configuration information (e.g., the ID of associated performance monitoring CSI configuration information).

[0279] The number of reference signals corresponding to the CSI configuration information refers to the number of reference signals included in the reference signal resource configuration associated with the CSI configuration information for channel measurement or interference measurement. The reference signals can be signals such as CSI-RS.

[0280] In some embodiments, high-level parameter information associated with CSI configuration information may include parameters for indicating the configuration of prediction windows and / or observation windows for CSI prediction, parameters for the temporal information used in CSI compression, specifying the AI ​​model ID, dataset ID, etc., as well as other AI use cases (such as beam management, CSI compression, CSI prediction) or purposes (inference, model monitoring, data collection) for determining the CSI transmission.

[0281] In some embodiments, the priority order of CSI configuration information can be determined based on the index number of the serving cell and the identifier of the CSI configuration information.

[0282] The following example illustrates the priority of CSI configuration information.

[0283] For any two CSI configuration information sets, event-triggered CSI transmission takes precedence over aperiodic CSI transmission, aperiodic CSI transmission takes precedence over semi-persistent CSI transmission, and semi-persistent CSI transmission takes precedence over periodic CSI transmission. If the priority of two CSI configuration information sets cannot be determined using the above information, the feedback quantity includes L1-SINR or L1-RSRP, with priority given to either. If the priority of two CSI configuration information sets still cannot be determined, the feedback quantity includes Squired Generalized Cosine Similarity (SGCS) and Normalized Mean Squared Error (MSE). Error (NMSE) or CSI configuration information indicating parameters used for performance monitoring of AI models takes priority; if the priority of two CSI configuration information cannot be determined, then CSI configuration information including other feedback quantities takes priority. These other feedback quantities may include parameters or higher-level parameters indicating that the CSI transmission is of the first type, and parameters or higher-level parameters indicating that the CSI transmission is of the second type. The first type of transmission can be based on non-AI CSI transmission, and the second type of transmission can be based on AI CSI transmission. When determining priority based on other feedback quantities, CSI configuration information indicating parameters of the first type of transmission takes priority; if the priority of two CSI configuration information cannot be determined... Priority is given to CSI configuration information that includes parameters for CSI compression data collection on the network device side. These parameters include parameters indicating data collection for spatiotemporal frequency domain CSI compression and parameters indicating data collection for spatial frequency domain CSI compression. When determining priority based on these parameters, CSI configuration information that includes parameters indicating data collection for spatiotemporal frequency domain CSI compression takes precedence. If the priority of two CSI configuration information cannot be determined, the CSI configuration information with the smaller serving cell sequence number takes precedence. If the priority of two CSI configuration information cannot be determined either, the CSI configuration information with the lower ID takes precedence.

[0284] The IDs for CSI configuration information can be determined by the network device. For example, the network device can determine that IDs used for beam management belong to range A, where non-AI beam management belongs to A1 and AI beam management belongs to A2; IDs used for other CSI transmissions (not data collection or performance monitoring) belong to range B, where first-type CSI transmissions belong to B1 and second-type CSI transmissions belong to B2; IDs used for performance monitoring CSI transmissions belong to range C, IDs based on spatiotemporal frequency domain CSI transmissions belong to C1, and IDs based on spatial frequency domain CSI transmissions belong to C2; and IDs used for AI data collection belong to range D.

[0285] In some embodiments, the network device may determine the ID range in different ways. The purpose of CSI reporting reflected in the ID can be determined by the range or by the value of the bits (e.g., the last bit of purpose A is 1, and the last two bits of purpose B are 10). This disclosure does not limit this.

[0286] In some embodiments, the CSI configuration information in the CSI configuration information set is determined in descending order of priority, and when calculating all the CSI configuration information in the CSI configuration information set, the CPU usage is less than the CPU limit, and the target resource usage is less than the target resource limit. The terminal device can determine the CSI configuration information set according to the following two feasible implementation methods:

[0287] One feasible implementation method:

[0288] The terminal device can select CSI configuration information in order of priority from multiple CSI configuration information. After each selection of CSI configuration information, the terminal device recalculates the CPU and target resources occupied by the selected CSI configuration information. If the CPU occupied by the selected CSI configuration information exceeds the CPU limit or the target resources occupied exceed the target resource limit after selecting a CSI configuration information, the terminal device can determine the CSI configuration information set before the selected CSI configuration information.

[0289] For example, the priority order of CSI configuration information is: CSI configuration information 1 - CSI configuration information 2 - CSI configuration information 3 - CSI configuration information 4. The terminal device selects CSI configuration information in sequence. When the terminal device selects CSI configuration information 4, the target resources occupied by CSI configuration information 1, CSI configuration information 2, CSI configuration information 3, and CSI configuration information 4 exceed the upper limit of the target resources. Then the terminal device can determine that the CSI configuration information set includes CSI configuration information 1, CSI configuration information 2, and CSI configuration information 3.

[0290] In some embodiments, once the CSI configuration information occupies any type of target resource exceeding the upper limit, the terminal device determines that the occupied target resource exceeds the upper limit of the target resource. For example, if the computing resources occupied by the CSI configuration information exceed the upper limit, but the other resources occupied do not exceed the upper limit, the terminal device determines that the target resources occupied by the CSI configuration information exceed the upper limit of the target resources; if the cache resources occupied by the CSI configuration information exceed the upper limit, but the other resources occupied do not exceed the upper limit, the terminal device determines that the target resources occupied by the CSI configuration information exceed the upper limit of the target resources; if the memory resources occupied by the CSI configuration information exceed the upper limit of the target resources, but the other resources occupied do not exceed the upper limit, the terminal device determines that the target resources occupied by the CSI configuration information exceed the upper limit of the target resources; if the storage space resources occupied by the CSI configuration information exceed the upper limit, but the other resources occupied do not exceed the upper limit, the terminal device determines that the target resources occupied by the CSI configuration information exceed the upper limit of the target resources; when the computing resources used for AI data processing occupied by the target resources do not exceed the upper limit of computing resources, the memory resources occupied do not exceed the upper limit of memory resources, the cache resources occupied do not exceed the upper limit of cache resources, and the storage space resources occupied do not exceed the upper limit of storage space resources, the terminal device determines that the target resources do not exceed the upper limit of the target resources.

[0291] Another feasible implementation method:

[0292] The terminal device can select CSI configuration information in order of priority from multiple CSI configuration information. After each selection, the terminal device recalculates the CPU and target resources occupied by the selected CSI configuration information. If the CPU occupied by the selected CSI configuration information exceeds the CPU limit or the target resources occupied exceed the target resource limit after selection, the terminal device will not select that CSI configuration information and will continue to select CSI configuration information in order of priority. The terminal device can then define the selected CSI configuration information as a set of CSI configuration information.

[0293] For example, the priority order of CSI configuration information is: CSI configuration information 1 - CSI configuration information 2 - CSI configuration information 3 - CSI configuration information 4. The terminal device selects CSI configuration information in sequence. When the terminal device selects CSI configuration information 3, if the target resources occupied by CSI configuration information 1, CSI configuration information 2, and CSI configuration information 3 exceed the upper limit of the target resources, the terminal device may choose not to select CSI configuration information 3 and instead select CSI configuration information 4. When the target resources occupied by CSI configuration information 1, CSI configuration information 2, and CSI configuration information 4 do not exceed the upper limit of the target resources and the CPU occupied does not exceed the upper limit of the CPU, the terminal device determines that the CSI configuration information set includes CSI configuration information 1, CSI configuration information 2, and CSI configuration information 4.

[0294] It should be noted that when a network device receives CSI, it can also determine the priority of the CSI configuration information according to a pre-agreed priority order, and this embodiment does not limit this.

[0295] S602. Determine the target CSI configuration information from the CSI configuration information set.

[0296] In some embodiments, if the physical channels carrying two CSI transmissions overlap by at least one Orthogonal Frequency Division Multiplexing (OFDM) symbol in time, the lower-priority CSI is not transmitted. Therefore, the terminal device can filter the CSI configuration information in the CSI configuration information set to obtain the target CSI configuration information.

[0297] For example, the CSI configuration information set includes CSI configuration information 1, CSI configuration information 2, CSI configuration information 3, and CSI configuration information 4. If CSI configuration information 1, CSI configuration information 2, CSI configuration information 3, and CSI configuration information 4 do not overlap in their CSI transmission times, the terminal device can determine CSI configuration information 1, CSI configuration information 2, CSI configuration information 3, and CSI configuration information 4 as the target CSI configuration information. If the physical channels carrying the CSIs corresponding to CSI configuration information 3 and CSI configuration information 4 overlap by at least one OFDM symbol in their CSI transmission times, and the priority of CSI configuration information 3 is higher than the priority of CSI configuration information 4, the terminal device will abandon the transmission of the CSI corresponding to CSI configuration information 4, and the terminal device will determine CSI configuration information 1, CSI configuration information 2, and CSI configuration information 3 as the target CSI configuration information.

[0298] This disclosure provides a method for determining target CSI configuration information. The terminal device determines a set of CSI configuration information according to priority order. The CSI configuration information in the set is determined in descending order of priority. Furthermore, when calculating all CSI configuration information in the set, if the CPU usage is less than the CPU limit and the target resource usage is less than the target resource limit, the terminal device can determine the target CSI configuration information from the set. In this method, when target resources and / or CPU are insufficient, the terminal device can determine the priority of CSI configuration information based on multiple pieces of information, improving the accuracy of priority determination. Moreover, the terminal device can select the target CSI configuration information from multiple CSI configuration information according to priority order, improving the accuracy of the target CSI configuration information and thus improving the accuracy of CSI transmission.

[0299] Based on any of the above embodiments, CSI transmission may also include AI-based CSI transmission and non-AI-based CSI transmission. Furthermore, under certain conditions, AI-based CSI transmission may be switched to non-AI-based CSI transmission. The method for switching CSI transmission will be described below with reference to Figure 7.

[0300] Figure 7 is a schematic diagram of a method for switching CSI transmission provided in an embodiment of this disclosure. Referring to Figure 7, the method flow includes:

[0301] S701. Determine whether the AI-based CSI transmission meets the handover conditions.

[0302] The switching conditions include at least one of the following:

[0303] The number of AI-based CSI transmission failures exceeded the preset number;

[0304] AI-based CSI transmission is not being sent due to priority considerations.

[0305] The network device did not receive the AI-based CSI transmission.

[0306] S702. If the AI-based CSI transmission meets the switching conditions, switch the AI-based CSI transmission to a non-AI-based CSI transmission.

[0307] In some embodiments, the failure of AI-based CSI transmission may be due to insufficient capabilities of the terminal device (e.g., AI model malfunction or insufficient computing power to use the AI ​​model), preventing the transmission of AI-based CSI. When the number of failed AI-based CSI transmissions exceeds a preset number, the terminal device can switch from AI-based CSI transmission to non-AI-based CSI transmission.

[0308] For example, if a terminal device fails to send AI-based CSI for N consecutive times (an integer greater than 0), or sends fewer than N AI-based CSIs in N consecutive transmission opportunities, the terminal device can switch from sending AI-based CSIs to sending non-AI-based CSIs.

[0309] In some embodiments, the terminal device sends AI-based CSI; however, if the network device fails to receive or repeatedly fails to receive AI-based CSI transmissions (e.g., network device failure), the terminal device can switch from AI-based CSI transmission to non-AI-based CSI transmission.

[0310] In some embodiments, AI-based CSI transmission is not sent due to priority reasons, and the terminal device can switch from AI-based CSI transmission to non-AI-based CSI transmission.

[0311] For example, if AI-based CSI transmission is on the same physical channel as other CSI transmissions and there are overlapping OFDM symbols in time, the priority of AI-based CSI transmission is lower than that of other CSI transmissions. The terminal device will not transmit AI-based CSI, and the terminal device can switch from AI-based CSI transmission to non-AI-based CSI transmission.

[0312] For example, if the target resources and / or CPU are insufficient and AI-based CSI transmission has a low priority, the terminal device will not transmit AI-based CSI, and the terminal device can switch from AI-based CSI transmission to non-AI-based CSI transmission.

[0313] In some embodiments, the handover conditions can be predefined or determined by the network device. For example, the handover conditions can be pre-agreed upon or determined by the network device. After determining the handover conditions, the network device can send the handover conditions to the terminal device.

[0314] In some embodiments, whether AI-based CSI transmission meets the handover conditions is determined by the terminal device or network device.

[0315] When the terminal device determines whether AI-based CSI transmission meets the handover conditions, switching from AI-based CSI transmission to non-AI-based CSI transmission includes: sending a handover request to the network device and receiving first indication information sent by the network device.

[0316] The first indication information can be used to indicate whether to switch from AI-based CSI transmission to non-AI-based CSI transmission. For example, if the terminal device determines that AI-based CSI transmission meets the handover conditions, the terminal device can send a handover request to the network device. The network device can determine whether the terminal device can switch from AI-based CSI transmission to non-AI-based CSI transmission. If yes, the first indication information sent by the network device to the terminal device can instruct the terminal device to switch from AI-based CSI transmission to non-AI-based CSI transmission; if no, the first indication information sent by the network device to the terminal device can instruct the terminal device not to perform the handover.

[0317] In some embodiments, when the terminal device determines that AI-based CSI transmission meets the switching conditions, the terminal device may also automatically switch AI-based CSI transmission to non-AI-based CSI transmission.

[0318] When the switching conditions for AI-based CSI transmission are determined by the network device, switching from AI-based CSI transmission to non-AI-based CSI transmission includes: receiving second indication information sent by the network device.

[0319] The second indication information can be used to instruct the switching of AI-based CSI transmission to non-AI-based CSI transmission. For example, when the network device determines that the switching conditions for AI-based CSI transmission are met and the terminal device can switch, the network device can send the second indication information to the terminal device. After receiving the second indication information, the terminal device can switch the AI-based CSI transmission to non-AI-based CSI transmission.

[0320] In some embodiments, non-AI-based CSI transmission may be pre-configured by the network device (e.g., predefined by the network device, or configured as non-AI-based CSI transmission when the network device is configured to use AI-based CSI transmission), or it may be additionally configured by the network device when the AI-based CSI transmission meets the switching conditions. This disclosure does not limit this aspect.

[0321] This disclosure provides a method for switching CSI transmission. A terminal device determines whether AI-based CSI transmission meets switching conditions. If the AI-based CSI transmission meets the switching conditions, the terminal device can switch from AI-based CSI transmission to non-AI-based CSI transmission. In this method, when the number of failed AI-based CSI transmissions exceeds a preset number, AI-based CSI transmissions are not sent due to priority reasons, or the network device does not receive the AI-based CSI transmission, the terminal device can promptly switch from AI-based CSI transmission to non-AI-based CSI transmission, thereby improving the success rate of CSI transmission.

[0322] Based on any of the above embodiments, the channel state information processing method may further include the network device sending second information to the terminal device to activate two CSI configuration information. The second information for activating two CSI configuration information will be described below with reference to FIG8.

[0323] Figure 8 is a schematic diagram of a method for sending second information according to an embodiment of this disclosure. Referring to Figure 8, the method flow includes:

[0324] S801, The network device sends the second information to the terminal device.

[0325] The second information is used to activate and / or trigger the first CSI configuration information and the second CSI configuration information.

[0326] The first CSI configuration information can be used for AI-based CSI transmission, while the second CSI configuration information can be used for performance monitoring of the first CSI configuration information (the AI ​​model of the first CSI configuration information). For example, when the first CSI configuration information is activated, the terminal device can perform inference and / or prediction based on the AI ​​model; when the second CSI configuration information is activated, the terminal device can monitor the performance of the inference and / or prediction based on the AI ​​model. For example, the first CSI configuration information is used for CSI spatial frequency domain compression and feedback of RI, CQI, and PMI; the second CSI configuration information can monitor the performance of the first CSI configuration information.

[0327] In some embodiments, the second information includes the identifier of the first CSI configuration information and the identifier of the second CSI configuration information.

[0328] In some embodiments, the second configuration information includes an identifier of the first CSI configuration information, and the first CSI configuration information may indicate the second CSI configuration information, or the timeline between the first CSI configuration information and the second CSI configuration information may be associated.

[0329] For example, the second information may include an identifier of the first CSI configuration information, and the second information may also include an identifier of the second CSI configuration information, so that the second information can activate both the first and second CSI configuration information simultaneously.

[0330] For example, the second information may include an identifier of the first CSI configuration information, and the first CSI configuration information may include an identifier of the second CSI configuration information, so that the second information can activate both the first CSI configuration information and the second CSI configuration information simultaneously.

[0331] For example, the second information may include an identifier of the first CSI configuration information, and the first CSI configuration information and the second CSI configuration information are associated with a timeline, so that the second information can activate both the first CSI configuration information and the second CSI configuration information simultaneously.

[0332] In some embodiments, the second information includes an identifier of the second configuration information, and the second CSI configuration information may indicate the first CSI configuration information, or the second CSI configuration information may be associated with a timeline between the first CSI configuration information and the second CSI configuration information.

[0333] For example, the second information may include an identifier of the second CSI configuration information, and the second CSI configuration information may include an identifier of the first CSI configuration information, so that the second information can activate both the first CSI configuration information and the second CSI configuration information simultaneously.

[0334] For example, the second information may include an identifier of the second CSI configuration information, and the second CSI configuration information is associated with the timeline of the first CSI configuration information, so that the second information can activate both the first and second CSI configuration information simultaneously.

[0335] In some embodiments, timeline association may be configured or predefined separately by the network device for the first CSI configuration information and / or the second CSI configuration information.

[0336] In some embodiments, timeline correlation can be expressed as follows: the timeline requirement of the first CSI configuration information is the timeline requirement of the second CSI reporting configuration plus K; or, the timeline requirement of the second CSI configuration information is the timeline requirement of the first CSI configuration information plus K. Here, K can be a single value or multiple values ​​related to the CSI configuration information (e.g., the prediction time and number of prediction times for AI-based CSI prediction). K can be a fixed value or a calculated value based on the capabilities of the terminal device.

[0337] For example, the timeline correlation can be the time difference between the earliest feedback time (such as uplink symbol or time slot) of the i-th (greater than 0 integer) feedback of the first CSI configuration information and the earliest feedback time of the p-th (greater than 0 integer) feedback of the second CSI configuration information. This time difference can be one or more values, which can be predefined, configured by the network device, or associated with the capabilities of the terminal device.

[0338] S802, The terminal device sends the CSI corresponding to the first configuration information and the CSI corresponding to the second configuration information to the network device.

[0339] The terminal device can determine the activated first CSI configuration information and second CSI configuration information based on the second information, determine the CSI corresponding to the first configuration information based on the first CSI configuration information, determine the CSI corresponding to the first configuration information based on the second CSI configuration information, and send the CSI corresponding to the first configuration information and the CSI corresponding to the second configuration information to the network device.

[0340] This disclosure provides a method for sending second information. A network device sends the second information to a terminal device, and the terminal device sends a CSI corresponding to first configuration information and a CSI corresponding to second configuration information to the network device. In this method, since the second information can activate and / or trigger the first and second CSI configuration information, the network device can activate both CSI configuration information based on a single signaling signal and clearly define the association between the two CSI configuration information systems, thereby saving communication resources and improving communication efficiency.

[0341] Figure 9 is a schematic diagram of a channel state information processing device provided in an embodiment of this disclosure. Referring to Figure 9, the channel state information processing device 900 includes a receiving module 901 and a transmitting module 902, wherein:

[0342] The receiving module 901 is used to receive first information sent by the network device, the first information being used to trigger and / or activate Channel State Information (CSI) configuration information.

[0343] The sending module 902 is used to send the CSI to the network device. The CSI is sent based on the CSI processing unit CPU, the target resource, and the CSI configuration information. The target resource is associated with the AI ​​processing of the CSI.

[0344] In some embodiments, the target resource includes at least one of the following:

[0345] Computing resources used for AI data processing;

[0346] Memory resources used for AI data processing;

[0347] Cache resources used for AI data processing;

[0348] Storage space resources used for AI data processing.

[0349] In some embodiments, the type of AI data processing includes at least one of the following:

[0350] Based on the AI ​​data, perform inference and / or prediction using the AI ​​model;

[0351] Based on the AI ​​data, perform AI model training tasks or related data collection;

[0352] Monitor the performance of AI model inference and / or prediction.

[0353] In some embodiments, the sending module is used to:

[0354] In the triggered and / or activated CSI configuration information, target CSI configuration information is determined, which is determined based on the target resources and / or CPU occupied by each CSI configuration information, and the target time for occupying the target resources and / or CPU.

[0355] The CSI is sent to the network device according to the target CSI configuration information.

[0356] In some embodiments, the target resources and / or CPU occupied by the CSI configuration information are determined based on at least one of the following:

[0357] The feedback quantity in the CSI configuration information;

[0358] High-level parameters associated with the CSI configuration information;

[0359] The parameters in the CSI configuration information;

[0360] The reference signal resources corresponding to the CSI configuration information;

[0361] Indicators or associated high-level parameters used to indicate the capabilities of terminal devices.

[0362] In some embodiments, the start time of the target time is determined based on at least one of the following:

[0363] The first symbol is the first symbol that triggers and / or activates the CSI configuration information, the last symbol, or the symbol following the last symbol.

[0364] The second symbol is the first symbol, the last symbol, or the symbol following the last symbol of the first reference signal transmission timing corresponding to the current CSI.

[0365] The third symbol is the first symbol, the last symbol, or the symbol following the last symbol of the last reference signal transmission timing corresponding to the current CSI transmission.

[0366] The fourth symbol is either the first symbol after the measurement at the last reference signal transmission timing is completed, or the first symbol after the measurement result is processed;

[0367] The fifth symbol is either the first symbol after completing AI model inference and / or prediction, or the Mth symbol after completing the measurement of the reference signal transmission timing, where M is associated with the terminal device capability.

[0368] In some embodiments, the end time of the target time is determined based on at least one of the following:

[0369] The fourth symbol;

[0370] The fifth symbol;

[0371] The sixth symbol is either the first or last symbol of the current CSI transmission.

[0372] The seventh symbol is either the first or last symbol corresponding to the CSI transmission period indicated by the CSI configuration information;

[0373] The eighth symbol is either the first or last symbol of the deactivation signaling corresponding to the CSI configuration information.

[0374] In some embodiments, if the CPU and / or the target resources are insufficient, the target CSI configuration information is determined according to the priority order of the CPU and / or the target resources and the CSI configuration information.

[0375] In some embodiments, the priority order of the CSI configuration information is determined according to at least one of the following:

[0376] The periodic information in the CSI configuration information;

[0377] The CSI configuration information specifies the triggering method for sending CSI.

[0378] The feedback quantity in the CSI configuration information;

[0379] Information in the CSI configuration information;

[0380] The number of reference signals corresponding to the CSI configuration information;

[0381] High-level parameter information associated with the CSI configuration information;

[0382] Serving the community;

[0383] The identifier of the CSI configuration information.

[0384] In some embodiments, the sending module is used to:

[0385] According to the priority order, a CSI configuration information set is determined. The CSI configuration information in the CSI configuration information set is determined in descending order of priority. Furthermore, when calculating all the CSI configuration information in the CSI configuration information set, the CPU usage is less than the upper limit of the CPU, and the target resources used are less than the upper limit of the target resources.

[0386] The target CSI configuration information is determined from the CSI configuration information set.

[0387] In some embodiments, the sending module is further configured to:

[0388] Send the CPU limit and / or the target resource limit to the network device.

[0389] In some embodiments, the channel state information processing device 900 further includes a processing module 903, the processing module 903 being configured to:

[0390] Determine whether the AI-based CSI transmission meets the handover conditions;

[0391] If the AI-based CSI transmission meets the switching conditions, the AI-based CSI transmission will be switched to the non-AI-based CSI transmission.

[0392] In some embodiments, the switching conditions include at least one of the following:

[0393] The number of AI-based CSI transmission failures is greater than a preset number;

[0394] The AI-based CSI transmission is not sent due to priority considerations;

[0395] The network device did not receive the AI-based CSI transmission.

[0396] In some embodiments, the handover conditions are predefined or determined by the network device, and whether the AI-based CSI transmission satisfies the handover conditions is determined by the terminal device or the network device.

[0397] In some embodiments, the sending module is further configured to:

[0398] Send a handover request to the network device;

[0399] The system receives a first indication message sent by a network device, the first indication message being used to indicate whether to switch the AI-based CSI transmission to the non-AI-based CSI transmission.

[0400] In some embodiments, the receiving module is further configured to:

[0401] The system receives a second indication message sent by a network device, the second indication message being used to indicate switching the AI-based CSI transmission to the non-AI-based CSI transmission.

[0402] In some embodiments, the receiving module is further configured to:

[0403] The system receives second information sent by a network device. The second information is used to activate and / or trigger first CSI configuration information and second CSI configuration information. The first CSI configuration information is used for AI-based CSI transmission, and the second CSI configuration information is used for performance monitoring of the first CSI configuration information.

[0404] In some embodiments, the second information includes an identifier of the first CSI configuration information, and the first CSI configuration information indicates the second CSI configuration information, or the first CSI configuration information and the second CSI configuration information are associated with a timeline;

[0405] The second information includes an identifier of the second CSI configuration information, and the second CSI configuration information indicates the first CSI configuration information, or the second CSI configuration information is associated with a timeline between the first CSI configuration information and the second CSI configuration information.

[0406] Figure 10 is a schematic diagram of another channel state information processing device provided in an embodiment of this disclosure. Referring to Figure 10, the channel state information processing device 1000 includes a transmitting module 1001 and a receiving module 1002, wherein:

[0407] The sending module is used to send first information to the terminal device, the first information being used to trigger and / or activate Channel State Information (CSI) configuration information;

[0408] The receiving module is used to receive CSI sent by the terminal device. The CSI is sent based on the CPU, target resources, and CSI configuration information. The target resources are associated with the AI ​​processing of the CSI.

[0409] In some embodiments, the target resource includes at least one of the following:

[0410] Computing resources used for AI data processing;

[0411] Memory resources used for AI data processing;

[0412] Cache resources used for AI data processing;

[0413] Storage space resources used for AI data processing.

[0414] In some embodiments, the type of AI data processing includes at least one of the following:

[0415] Based on the AI ​​data, perform inference and / or prediction using the AI ​​model;

[0416] Based on the AI ​​data, perform AI model training tasks or related data collection;

[0417] Monitor the performance of AI model inference and / or prediction.

[0418] In some embodiments, the receiving module is used to:

[0419] According to the target CSI configuration information, the terminal device sends CSIs. The target CSI configuration information is determined based on the target resources and / or CPU occupied by each CSI configuration information, and the target time for occupying the target resources and / or CPU.

[0420] In some embodiments, the target resources and / or CPU occupied by the CSI configuration information are determined based on at least one of the following:

[0421] The feedback quantity in the CSI configuration information;

[0422] High-level parameters associated with the CSI configuration information;

[0423] The parameters in the CSI configuration information;

[0424] The reference signal resources corresponding to the CSI configuration information;

[0425] Indicators or associated high-level parameters used to indicate the capabilities of terminal devices.

[0426] In some embodiments, the start time of the target time is determined based on at least one of the following:

[0427] The first symbol is the first symbol, the last symbol, or the symbol following the last symbol after triggering and / or activating the CSI configuration information;

[0428] The second symbol is the first symbol, the last symbol, or the symbol following the last symbol of the first reference signal transmission timing corresponding to the current CSI.

[0429] The third symbol is the first symbol, the last symbol, or the symbol following the last symbol of the last reference signal transmission timing corresponding to the current CSI.

[0430] The fourth symbol is either the first symbol after the measurement at the last reference signal transmission timing is completed, or the first symbol after the measurement result is processed;

[0431] The fifth symbol is either the first symbol after completing AI model inference and / or prediction, or the Mth symbol after completing the measurement of the reference signal transmission timing, where M is associated with the capabilities of the terminal device.

[0432] In some embodiments, the end time of the target time is determined based on at least one of the following:

[0433] The fourth symbol;

[0434] The fifth symbol;

[0435] The sixth symbol is either the first or last symbol received in the current CSI.

[0436] The seventh symbol is either the first or last symbol corresponding to the CSI transmission period indicated by the CSI configuration information;

[0437] The eighth symbol is either the first or last symbol of the deactivation signaling corresponding to the CSI configuration information.

[0438] In some embodiments, if the CPU and / or the target resources are insufficient, the target CSI configuration information is determined according to the priority order of the CPU and / or the target resources and the CSI configuration information.

[0439] In some embodiments, the priority order of the CSI configuration information is determined according to at least one of the following:

[0440] The periodic information in the CSI configuration information;

[0441] The CSI configuration information specifies the triggering method for sending CSI.

[0442] The feedback quantity in the CSI configuration information;

[0443] Information in the CSI configuration information;

[0444] The number of reference signals corresponding to the CSI configuration information;

[0445] High-level parameter information associated with the CSI configuration information;

[0446] Serving the community;

[0447] The identifier of the CSI configuration information.

[0448] In some embodiments, the receiving module is further configured to:

[0449] The upper limit of CPU and / or the upper limit of target resources sent by the receiving terminal device.

[0450] In some embodiments, the channel state information processing device 1000 further includes a processing module 1003, the processing module 1003 being configured to:

[0451] Determine whether the AI-based CSI transmission meets the handover conditions;

[0452] If the AI-based CSI transmission meets the switching conditions, the AI-based CSI transmission will be switched to the non-AI-based CSI transmission.

[0453] In some embodiments, the switching conditions include at least one of the following:

[0454] The number of AI-based CSI transmission failures is greater than a preset number;

[0455] The AI-based CSI transmission is not sent due to priority considerations;

[0456] The network device did not receive the AI-based CSI transmission.

[0457] In some embodiments, the handover conditions are predefined or determined by the network device, and whether the AI-based CSI transmission satisfies the handover conditions is determined by the terminal device or the network device.

[0458] In some embodiments, the receiving module is further configured to:

[0459] Receive handover requests sent by terminal devices;

[0460] Send a first indication message to the terminal device, the first indication message being used to indicate whether to switch the AI-based CSI transmission to the non-AI-based CSI transmission.

[0461] In some embodiments, the sending module is further configured to:

[0462] Send a second instruction to the terminal device, the second instruction being used to instruct the switching of the AI-based CSI transmission to the non-AI-based CSI transmission.

[0463] In some embodiments, the sending module is further configured to:

[0464] Send a second message to the terminal device. The second message is used to activate and / or trigger a first CSI configuration message and a second CSI configuration message. The first CSI configuration message is used for AI-based CSI transmission, and the second CSI configuration message is used for performance monitoring of the first CSI configuration message.

[0465] In some embodiments, the second information includes an identifier of the first CSI configuration information, and the first CSI configuration information indicates the second CSI configuration information, or the first CSI configuration information and the second CSI configuration information are associated with a timeline;

[0466] The second information includes an identifier of the second CSI configuration information, and the second CSI configuration information indicates the first CSI configuration information, or the second CSI configuration information is associated with a timeline between the first CSI configuration information and the second CSI configuration information.

[0467] It should be noted that the division of units in the embodiments of this disclosure is illustrative and only represents one logical functional division. In actual implementation, other division methods may be used. Furthermore, the functional units in the various embodiments of this disclosure 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. The integrated units described above can be implemented in hardware or as software functional units.

[0468] If the aforementioned integrated units are implemented as software functional units and sold or used as independent products, they can be stored in a processor-readable storage medium. Based on this understanding, the technical solution of this disclosure, in essence, or the part that contributes to related technologies, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) or processor to execute all or part of the steps of the methods of the various embodiments of this disclosure. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0469] It should be noted that the apparatus provided in this disclosure can implement all the method steps implemented in the above method embodiments and can achieve the same technical effect. Therefore, the parts and beneficial effects that are the same as those in the method embodiments will not be described again here.

[0470] Figure 11 is a schematic diagram of a terminal device provided in an embodiment of this disclosure. Referring to Figure 11, the terminal device includes a memory 1110, a transceiver 1120, and a processor 1130.

[0471] The memory 1110 is used to store computer programs;

[0472] The transceiver 1120 is used to send and receive data under the control of the processor;

[0473] The processor 1130 is configured to read the computer program in the memory and perform the following operations:

[0474] Receive first information sent by the network device, the first information being used to trigger and / or activate Channel State Information (CSI) configuration information;

[0475] The CSI is sent to the network device. The CSI is sent based on the CSI processing unit CPU, the target resource, and the CSI configuration information. The target resource is associated with the AI ​​processing of the CSI.

[0476] In some embodiments, the target resource includes at least one of the following:

[0477] Computing resources used for AI data processing;

[0478] Memory resources used for AI data processing;

[0479] Cache resources used for AI data processing;

[0480] Storage space resources used for AI data processing.

[0481] In some embodiments, the type of AI data processing includes at least one of the following:

[0482] Based on the AI ​​data, perform inference and / or prediction using the AI ​​model;

[0483] Based on the AI ​​data, perform AI model training tasks or related data collection;

[0484] Monitor the performance of AI model inference and / or prediction.

[0485] In some embodiments, sending the CSI to the network device includes:

[0486] In the triggered and / or activated CSI configuration information, target CSI configuration information is determined, which is determined based on the target resources and / or CPU occupied by each CSI configuration information, and the target time for occupying the target resources and / or CPU.

[0487] The CSI is sent to the network device according to the target CSI configuration information.

[0488] In some embodiments, the target resources and / or CPU occupied by the CSI configuration information are determined based on at least one of the following:

[0489] The feedback quantity in the CSI configuration information;

[0490] High-level parameters associated with the CSI configuration information;

[0491] The parameters in the CSI configuration information;

[0492] The reference signal resources corresponding to the CSI configuration information;

[0493] Indicators or associated high-level parameters used to indicate the capabilities of terminal devices.

[0494] In some embodiments, the start time of the target time is determined based on at least one of the following:

[0495] The first symbol is the first symbol that triggers and / or activates the CSI configuration information, the last symbol, or the symbol following the last symbol.

[0496] The second symbol is the first symbol, the last symbol, or the symbol following the last symbol of the first reference signal transmission timing corresponding to the current CSI.

[0497] The third symbol is the first symbol, the last symbol, or the symbol following the last symbol of the last reference signal transmission timing corresponding to the current CSI transmission.

[0498] The fourth symbol is either the first symbol after the measurement at the last reference signal transmission timing is completed, or the first symbol after the measurement result is processed;

[0499] The fifth symbol is either the first symbol after completing AI model inference and / or prediction, or the Mth symbol after completing the measurement of the reference signal transmission timing, where M is associated with the terminal device capability.

[0500] In some embodiments, the end time of the target time is determined based on at least one of the following:

[0501] The fourth symbol;

[0502] The fifth symbol;

[0503] The sixth symbol is either the first or last symbol of the current CSI transmission.

[0504] The seventh symbol is either the first or last symbol corresponding to the CSI transmission period indicated by the CSI configuration information;

[0505] The eighth symbol is either the first or last symbol of the deactivation signaling corresponding to the CSI configuration information.

[0506] In some embodiments, if the CPU and / or the target resources are insufficient, the target CSI configuration information is determined according to the priority order of the CPU and / or the target resources and the CSI configuration information.

[0507] In some embodiments, the priority order of the CSI configuration information is determined according to at least one of the following:

[0508] The periodic information in the CSI configuration information;

[0509] The CSI configuration information specifies the triggering method for sending CSI.

[0510] The feedback quantity in the CSI configuration information;

[0511] Information in the CSI configuration information;

[0512] The number of reference signals corresponding to the CSI configuration information;

[0513] High-level parameter information associated with the CSI configuration information;

[0514] Serving the community;

[0515] The identifier of the CSI configuration information.

[0516] In some embodiments, determining the target CSI configuration information based on the priority order of the CPU and / or the target resource, and the CSI configuration information, includes:

[0517] According to the priority order, a CSI configuration information set is determined. The CSI configuration information in the CSI configuration information set is determined in descending order of priority. Furthermore, when calculating all the CSI configuration information in the CSI configuration information set, the CPU usage is less than the upper limit of the CPU, and the target resources used are less than the upper limit of the target resources.

[0518] The target CSI configuration information is determined from the CSI configuration information set.

[0519] In some embodiments, the method further includes:

[0520] Send the CPU limit and / or the target resource limit to the network device.

[0521] In some embodiments, the CSI transmission includes AI-based CSI transmission and non-AI-based CSI transmission, and the method further includes:

[0522] Determine whether the AI-based CSI transmission meets the handover conditions;

[0523] If the AI-based CSI transmission meets the switching conditions, the AI-based CSI transmission will be switched to the non-AI-based CSI transmission.

[0524] In some embodiments, the switching conditions include at least one of the following:

[0525] The number of AI-based CSI transmission failures is greater than a preset number;

[0526] The AI-based CSI transmission is not sent due to priority considerations;

[0527] The network device did not receive the AI-based CSI transmission.

[0528] In some embodiments, the handover conditions are predefined or determined by the network device, and whether the AI-based CSI transmission satisfies the handover conditions is determined by the terminal device or the network device.

[0529] In some embodiments, when the switching condition for the AI-based CSI transmission is determined by the terminal device, switching the AI-based CSI transmission to the non-AI-based CSI transmission includes:

[0530] Send a handover request to the network device;

[0531] The system receives a first indication message sent by a network device, the first indication message being used to indicate whether to switch the AI-based CSI transmission to the non-AI-based CSI transmission.

[0532] In some embodiments, when the switching condition for the AI-based CSI transmission is determined by the network device, switching the AI-based CSI transmission to the non-AI-based CSI transmission includes:

[0533] The system receives a second indication message sent by a network device, the second indication message being used to indicate switching the AI-based CSI transmission to the non-AI-based CSI transmission.

[0534] In some embodiments, the method further includes:

[0535] The system receives second information sent by a network device. The second information is used to activate and / or trigger first CSI configuration information and second CSI configuration information. The first CSI configuration information is used for AI-based CSI transmission, and the second CSI configuration information is used for performance monitoring of the first CSI configuration information.

[0536] In some embodiments, the second information includes an identifier of the first CSI configuration information, and the first CSI configuration information indicates the second CSI configuration information, or the first CSI configuration information and the second CSI configuration information are associated with a timeline;

[0537] The second information includes an identifier of the second CSI configuration information, and the second CSI configuration information indicates the first CSI configuration information, or the second CSI configuration information is associated with a timeline between the first CSI configuration information and the second CSI configuration information.

[0538] In some embodiments, the terminal device may further include a user interface 1140. For different terminal devices, the user interface 1140 may also be an interface that can connect to external or internal devices, including but not limited to keypad, display, speaker, microphone, joystick, etc.

[0539] In Figure 11, the bus architecture may include any number of interconnected buses and bridges, linking various circuits of one or more processors represented by processor 1130 and memory represented by memory 1110. The bus architecture may also link various other circuits such as peripheral devices, voltage regulators, and power management circuits, which are well known in the art and therefore will not be described further herein. The bus interface provides an interface. Transceiver 1120 may be multiple elements, including transmitters and receivers, providing units for communicating with various other devices over transmission media, including wireless channels, wired channels, optical fibers, etc. Processor 1130 is responsible for managing the bus architecture and general processing, and memory 1110 may store data used by processor 1130 during operation.

[0540] Optionally, the processor 1130 may be a central processing unit (CPU), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or a complex programmable logic device (CPLD), and the processor may also adopt a multi-core architecture.

[0541] The processor 1130 executes any of the methods provided in the embodiments of this disclosure by calling a computer program stored in the memory 1110, according to the obtained executable instructions. The processor 1130 and the memory 1110 may also be physically separated.

[0542] It should be noted that the physical device provided in this disclosure can implement all the method steps implemented by the physical device in the above method embodiments and can achieve the same technical effect. Therefore, the parts and beneficial effects that are the same as those in the method embodiments will not be described again here.

[0543] Figure 12 is a schematic diagram of a network device provided in an embodiment of this disclosure. Referring to Figure 12, the network device includes a memory 1210, a transceiver 1220, and a processor 1230.

[0544] The memory 1210 is used to store computer programs;

[0545] The transceiver 1220 is used to send and receive data under the control of the processor;

[0546] The processor 1230 is configured to read the computer program in the memory and perform the following operations:

[0547] The first information sent to the terminal device is used to trigger and / or activate the Channel State Information (CSI) configuration information.

[0548] The terminal device receives a CSI sent by the terminal device. The CSI is sent based on the CPU, target resources, and CSI configuration information. The target resources are associated with the AI ​​processing of the CSI.

[0549] In some embodiments, the target resource includes at least one of the following:

[0550] Computing resources used for AI data processing;

[0551] Memory resources used for AI data processing;

[0552] Cache resources used for AI data processing;

[0553] Storage space resources used for AI data processing.

[0554] In some embodiments, the type of AI data processing includes at least one of the following:

[0555] Based on the AI ​​data, perform inference and / or prediction using the AI ​​model;

[0556] Based on the AI ​​data, perform AI model training tasks or related data collection;

[0557] Monitor the performance of AI model inference and / or prediction.

[0558] In some embodiments, receiving CSI sent by the terminal device includes:

[0559] According to the target CSI configuration information, the terminal device sends CSIs. The target CSI configuration information is determined based on the target resources and / or CPU occupied by each CSI configuration information, and the target time for occupying the target resources and / or CPU.

[0560] In some embodiments, the target resources and / or CPU occupied by the CSI configuration information are determined based on at least one of the following:

[0561] The feedback quantity in the CSI configuration information;

[0562] High-level parameters associated with the CSI configuration information;

[0563] The parameters in the CSI configuration information;

[0564] The reference signal resources corresponding to the CSI configuration information;

[0565] Indicators or associated high-level parameters used to indicate the capabilities of terminal devices.

[0566] In some embodiments, the start time of the target time is determined based on at least one of the following:

[0567] The first symbol is the first symbol, the last symbol, or the symbol following the last symbol after triggering and / or activating the CSI configuration information;

[0568] The second symbol is the first symbol, the last symbol, or the symbol following the last symbol of the first reference signal transmission timing corresponding to the current CSI.

[0569] The third symbol is the first symbol, the last symbol, or the symbol following the last symbol of the last reference signal transmission timing corresponding to the current CSI.

[0570] The fourth symbol is either the first symbol after the measurement at the last reference signal transmission timing is completed, or the first symbol after the measurement result is processed;

[0571] The fifth symbol is either the first symbol after completing AI model inference and / or prediction, or the Mth symbol after completing the measurement of the reference signal transmission timing, where M is associated with the capabilities of the terminal device.

[0572] In some embodiments, the end time of the target time is determined based on at least one of the following:

[0573] The fourth symbol;

[0574] The fifth symbol;

[0575] The sixth symbol is either the first or last symbol received in the current CSI.

[0576] The seventh symbol is either the first or last symbol corresponding to the CSI transmission period indicated by the CSI configuration information;

[0577] The eighth symbol is either the first or last symbol of the deactivation signaling corresponding to the CSI configuration information.

[0578] In some embodiments, if the CPU and / or the target resources are insufficient, the target CSI configuration information is determined according to the priority order of the CPU and / or the target resources and the CSI configuration information.

[0579] In some embodiments, the priority order of the CSI configuration information is determined according to at least one of the following:

[0580] The periodic information in the CSI configuration information;

[0581] The CSI configuration information specifies the triggering method for sending CSI.

[0582] The feedback quantity in the CSI configuration information;

[0583] Information in the CSI configuration information;

[0584] The number of reference signals corresponding to the CSI configuration information;

[0585] High-level parameter information associated with the CSI configuration information;

[0586] Serving the community;

[0587] The identifier of the CSI configuration information.

[0588] In some embodiments, the method further includes:

[0589] The upper limit of CPU and / or the upper limit of target resources sent by the receiving terminal device.

[0590] In some embodiments, the CSI transmission includes AI-based CSI transmission and non-AI-based CSI transmission, and the method further includes:

[0591] Determine whether the AI-based CSI transmission meets the handover conditions;

[0592] If the AI-based CSI transmission meets the switching conditions, the AI-based CSI transmission will be switched to the non-AI-based CSI transmission.

[0593] In some embodiments, the switching conditions include at least one of the following:

[0594] The number of AI-based CSI transmission failures is greater than a preset number;

[0595] The AI-based CSI transmission is not sent due to priority considerations;

[0596] The network device did not receive the AI-based CSI transmission.

[0597] In some embodiments, the handover conditions are predefined or determined by the network device, and whether the AI-based CSI transmission satisfies the handover conditions is determined by the terminal device or the network device.

[0598] In some embodiments, when the switching condition for the AI-based CSI transmission is determined by the terminal device, switching the AI-based CSI transmission to the non-AI-based CSI transmission includes:

[0599] Receive handover requests sent by terminal devices;

[0600] Send a first indication message to the terminal device, the first indication message being used to indicate whether to switch the AI-based CSI transmission to the non-AI-based CSI transmission.

[0601] In some embodiments, when the switching condition for the AI-based CSI transmission is determined by the network device, switching the AI-based CSI transmission to the non-AI-based CSI transmission includes:

[0602] A second indication message is sent to the terminal device, the second indication message being used to indicate switching the AI-based CSI transmission to the non-AI-based CSI transmission.

[0603] In some embodiments, the method further includes:

[0604] Send a second message to the terminal device. The second message is used to activate and / or trigger a first CSI configuration message and a second CSI configuration message. The first CSI configuration message is used for AI-based CSI transmission, and the second CSI configuration message is used for performance monitoring of the first CSI configuration message.

[0605] In some embodiments, the second information includes an identifier of the first CSI configuration information, and the first CSI configuration information indicates the second CSI configuration information, or the first CSI configuration information and the second CSI configuration information are associated with a timeline;

[0606] The second information includes an identifier of the second CSI configuration information, and the second CSI configuration information indicates the first CSI configuration information, or the second CSI configuration information is associated with a timeline between the first CSI configuration information and the second CSI configuration information.

[0607] This disclosure also provides a processor-readable storage medium storing a computer program for causing a processor to perform the method described in any of the above method embodiments.

[0608] Processor-readable storage media can be any available medium or data storage device that a computer can access, including but not limited to magnetic storage (such as floppy disks, hard disks, magnetic tapes, magneto-optical disks (MOs), etc.), optical storage (such as CDs, DVDs, BDs, HVDs, etc.), and semiconductor storage (such as ROMs, EPROMs, EEPROMs, non-volatile memory (NAND flash), solid-state drives (SSDs)).

[0609] This disclosure also provides a computer program product, including a computer program that, when executed by a processor, implements the method described in any of the above method embodiments.

[0610] Those skilled in the art will understand that embodiments of this disclosure can be provided as methods, systems, or computer program products. Therefore, this disclosure can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this disclosure can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage and optical storage) containing computer-usable program code.

[0611] This disclosure is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this disclosure. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer-executable instructions. These computer-executable instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in one or more flowchart illustrations and / or one or more block diagrams.

[0612] These processor-executable instructions may also be stored in a processor-readable memory that can instruct a computer or other programmable data processing device to operate in a particular manner, such that the instructions stored in the processor-readable memory produce an article of manufacture including instruction means that implement the functions specified in one or more flowcharts and / or one or more block diagrams.

[0613] These processor-executable instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, such that the instructions, which execute on the computer or other programmable apparatus, provide steps for implementing the functions specified in one or more flowcharts and / or one or more block diagrams.

[0614] Obviously, those skilled in the art can make various modifications and variations to this disclosure without departing from its spirit and scope. Therefore, if such modifications and variations fall within the scope of the claims of this disclosure and their equivalents, this disclosure is also intended to include such modifications and variations.

Claims

1. A channel state information processing method, wherein, Applied to a terminal device, the method includes: Receive first information sent by the network device, the first information being used to trigger and / or activate Channel State Information (CSI) configuration information; The CSI is sent to the network device. The CSI is sent based on the CSI processing unit CPU, the target resource, and the CSI configuration information. The target resource is associated with the AI ​​processing of the CSI.

2. The method according to claim 1, wherein, The target resource includes at least one of the following: Computing resources used for AI data processing; Memory resources used for AI data processing; Cache resources used for AI data processing; Storage space resources used for AI data processing.

3. The method according to claim 2, wherein, The types of AI data processing include at least one of the following: Based on the AI ​​data, perform inference and / or prediction using the AI ​​model; Based on the AI ​​data, perform AI model training tasks or related data collection; Monitor the performance of AI model inference and / or prediction.

4. The method according to any one of claims 1-3, wherein, Sending the CSI to the network device includes: In the triggered and / or activated CSI configuration information, target CSI configuration information is determined, which is determined based on the target resources and / or CPU occupied by each CSI configuration information, and the target time for occupying the target resources and / or CPU. The CSI is sent to the network device according to the target CSI configuration information.

5. The method according to claim 4, wherein, The target resources and / or CPU occupied by the CSI configuration information are determined based on at least one of the following: The feedback quantity in the CSI configuration information; High-level parameters associated with the CSI configuration information; The parameters in the CSI configuration information; The reference signal resources corresponding to the CSI configuration information; Indicators or associated high-level parameters used to indicate the capabilities of terminal devices.

6. The method according to claim 4, wherein, The start time of the target time is determined based on at least one of the following: The first symbol is the first symbol that triggers and / or activates the CSI configuration information, the last symbol, or the symbol following the last symbol. The second symbol is the first symbol, the last symbol, or the symbol following the last symbol of the first reference signal transmission timing corresponding to the current CSI. The third symbol is the first symbol, the last symbol, or the symbol following the last symbol of the last reference signal transmission timing corresponding to the current CSI transmission. The fourth symbol is either the first symbol after the measurement at the last reference signal transmission timing is completed, or the first symbol after the measurement result is processed; The fifth symbol is either the first symbol after completing AI model inference and / or prediction, or the Mth symbol after completing the measurement of the reference signal transmission timing, where M is associated with the terminal device capability.

7. The method according to claim 6, wherein, The end time of the target time is determined based on at least one of the following: The fourth symbol; The fifth symbol; The sixth symbol is either the first or last symbol of the current CSI transmission. The seventh symbol is either the first or last symbol corresponding to the CSI transmission period indicated by the CSI configuration information; The eighth symbol is either the first or last symbol of the deactivation signaling corresponding to the CSI configuration information.

8. The method according to claim 4, wherein, In the event of insufficient CPU and / or target resources, target CSI configuration information is determined based on the priority order of the CPU and / or target resources and the CSI configuration information.

9. The method according to claim 8, wherein, The priority order of the CSI configuration information is determined based on at least one of the following: The periodic information in the CSI configuration information; The CSI configuration information specifies the triggering method for sending CSI. The feedback quantity in the CSI configuration information; Information in the CSI configuration information; The number of reference signals corresponding to the CSI configuration information; High-level parameter information associated with the CSI configuration information; Serving the community; The identifier of the CSI configuration information.

10. The method according to claim 8 or 9, wherein, Based on the CPU and / or the target resource, and the priority order of the CSI configuration information, the target CSI configuration information is determined, including: According to the priority order, a CSI configuration information set is determined. The CSI configuration information in the CSI configuration information set is determined in descending order of priority. Furthermore, when calculating all the CSI configuration information in the CSI configuration information set, the CPU usage is less than the upper limit of the CPU, and the target resources used are less than the upper limit of the target resources. The target CSI configuration information is determined from the CSI configuration information set.

11. The method according to any one of claims 1-3, wherein, The method further includes: Send the CPU limit and / or the target resource limit to the network device.

12. The method according to any one of claims 1-3, wherein, The CSI transmission includes AI-based CSI transmission and non-AI-based CSI transmission, and the method further includes: Determine whether the AI-based CSI transmission meets the switching conditions; If the AI-based CSI transmission meets the switching conditions, the AI-based CSI transmission will be switched to the non-AI-based CSI transmission.

13. The method according to claim 12, wherein, The switching conditions include at least one of the following: The number of AI-based CSI transmission failures is greater than a preset number; The AI-based CSI transmission is not sent due to priority considerations; The network device did not receive the AI-based CSI transmission.

14. The method according to claim 12, wherein, The handover conditions are predefined or determined by the network device, and whether the AI-based CSI transmission meets the handover conditions is determined by the terminal device or the network device.

15. The method according to claim 14, wherein, When the switching conditions for the AI-based CSI transmission are determined by the terminal device, switching the AI-based CSI transmission to the non-AI-based CSI transmission includes: Send a handover request to the network device; The system receives a first indication message sent by a network device, the first indication message being used to indicate whether to switch the AI-based CSI transmission to the non-AI-based CSI transmission.

16. The method of claim 14, wherein, If the switching conditions for the AI-based CSI transmission are determined by the network device, switching the AI-based CSI transmission to the non-AI-based CSI transmission includes: The system receives a second indication message sent by a network device, the second indication message being used to indicate switching the AI-based CSI transmission to the non-AI-based CSI transmission.

17. The method according to any one of claims 1-3, wherein, The method further includes: The system receives second information sent by a network device. The second information is used to activate and / or trigger first CSI configuration information and second CSI configuration information. The first CSI configuration information is used for AI-based CSI transmission, and the second CSI configuration information is used for performance monitoring of the first CSI configuration information.

18. The method according to claim 17, wherein, The second information includes an identifier of the first CSI configuration information, and the first CSI configuration information indicates the second CSI configuration information, or the timeline between the first CSI configuration information and the second CSI configuration information is associated. The second information includes an identifier of the second CSI configuration information, and the second CSI configuration information indicates the first CSI configuration information, or the second CSI configuration information is associated with a timeline between the first CSI configuration information and the second CSI configuration information.

19. A channel state information processing method, wherein, Applied to network devices, the method includes: The first information sent to the terminal device is used to trigger and / or activate the Channel State Information (CSI) configuration information. The terminal device receives a CSI sent by the terminal device. The CSI is sent based on the CPU, target resources, and CSI configuration information. The target resources are associated with the AI ​​processing of the CSI.

20. The method according to claim 19, wherein, The target resource includes at least one of the following: Computing resources used for AI data processing; Memory resources used for AI data processing; Cache resources used for AI data processing; Storage space resources used for AI data processing.

21. The method according to claim 19, wherein, Receiving CSI sent by the terminal device includes: According to the target CSI configuration information, the terminal device sends CSIs. The target CSI configuration information is determined based on the target resources and / or CPU occupied by each CSI configuration information, and the target time for occupying the target resources and / or CPU.

22. The method according to claim 21, wherein, The start time of the target time is determined based on at least one of the following: The first symbol is the first symbol, the last symbol, or the symbol following the last symbol after triggering and / or activating the CSI configuration information; The second symbol is the first symbol, the last symbol, or the symbol following the last symbol of the first reference signal transmission timing corresponding to the current CSI. The third symbol is the first symbol, the last symbol, or the symbol following the last symbol of the last reference signal transmission timing corresponding to the current CSI. The fourth symbol is either the first symbol after the measurement at the last reference signal transmission timing is completed, or the first symbol after the measurement result is processed; The fifth symbol is either the first symbol after completing AI model inference and / or prediction, or the Mth symbol after completing the measurement of the reference signal transmission timing, where M is associated with the capabilities of the terminal device.

23. The method according to claim 22, wherein, The end time of the target time is determined based on at least one of the following: The fourth symbol; The fifth symbol; The sixth symbol is either the first or last symbol received in the current CSI. The seventh symbol is either the first or last symbol corresponding to the CSI transmission period indicated by the CSI configuration information; The eighth symbol is either the first or last symbol of the deactivation signaling corresponding to the CSI configuration information.

24. The method according to claim 22, wherein, In the event of insufficient CPU and / or target resources, target CSI configuration information is determined based on the priority order of the CPU and / or target resources and the CSI configuration information.

25. The method according to claim 24, wherein, The priority order of the CSI configuration information is determined based on at least one of the following: The periodic information in the CSI configuration information; The CSI configuration information specifies the triggering method for sending CSI. The feedback quantity in the CSI configuration information; Information in the CSI configuration information; The number of reference signals corresponding to the CSI configuration information; High-level parameter information associated with the CSI configuration information; Serving the community; The identifier of the CSI configuration information.

26. A channel state information processing apparatus, wherein, Applied to terminal devices, including a receiving module and a transmitting module, wherein: The receiving module is used to receive first information sent by the network device, the first information being used to trigger and / or activate Channel State Information (CSI) configuration information; The sending module is used to send the CSI to the network device. The CSI is sent based on the CSI processing unit CPU, the target resource, and the CSI configuration information. The target resource is associated with the AI ​​processing of the CSI.

27. A channel state information processing apparatus, wherein, Applied to network devices, including transmitting and receiving modules, wherein: The sending module is used to send first information to the terminal device, the first information being used to trigger and / or activate Channel State Information (CSI) configuration information; The receiving module is used to receive CSI sent by the terminal device. The CSI is sent based on the CPU, target resources, and CSI configuration information. The target resources are associated with the AI ​​processing of the CSI.

28. A terminal device, wherein, Includes memory, transceiver, and processor: The memory is used to store computer programs; The transceiver is used to send and receive data under the control of the processor; The processor is configured to read the computer program from the memory and perform the following operations: Receive first information sent by the network device, the first information being used to trigger and / or activate Channel State Information (CSI) configuration information; The CSI is sent to the network device. The CSI is sent based on the CSI processing unit CPU, the target resource, and the CSI configuration information. The target resource is associated with the AI ​​processing of the CSI.

29. A network device, wherein, Includes memory, transceiver, and processor: The memory is used to store computer programs; The transceiver is used to send and receive data under the control of the processor; The processor is configured to read the computer program from the memory and perform the following operations: The first information sent to the terminal device is used to trigger and / or activate the Channel State Information (CSI) configuration information. The terminal device receives a CSI sent by the terminal device. The CSI is sent based on the CPU, target resources, and CSI configuration information. The target resources are associated with the AI ​​processing of the CSI.

30. A processor-readable storage medium, wherein, The processor-readable storage medium stores a computer program that causes the processor to perform the method as described in any one of claims 1-18, or the method as described in any one of claims 19-25.

Images