Communication method and communication system

Abstract

The present application relates to the technical field of communications, and provides a communication method and a communication system. A terminal device acquires first channel state information (CSI) capability information of the terminal device. The first CSI capability information comprises the size of a neural processing unit (NPU) and the size of a first CSI processing unit (CPU). The first CPU comprises an occupied CPU used for determining a first CSI report, or the first CPU comprises an occupied CPU used for determining the first CSI report and a second CSI report. The first CSI report represents a CSI report based on an artificial intelligence (AI) function, and the second CSI report represents a CSI report not based on the AI function, that is, the second CSI report is a conventional CSI report. Then, the terminal device reports the first CSI capability information to a network device, so that the network device learns an AI function-related capability of the terminal device side.

Description

A communication method and communication system

[0001] This application claims priority to Chinese Patent Application No. 202510144626.8, filed on February 7, 2025, entitled "A Communication Method and Communication System", the entire contents of which are incorporated herein by reference. Technical Field

[0002] This application relates to the field of communication technology, and in particular to a communication method and communication system. Background Technology

[0003] With low-frequency communication resources becoming increasingly scarce, wireless communication has begun to develop higher-frequency communication. However, due to the high attenuation, low penetration, and easy absorption characteristics of high-frequency electromagnetic waves, communication equipment requires larger antennas to ensure communication quality. While large-scale antennas can improve signal strength, they also incur significant overhead. To reduce overhead, artificial intelligence (AI) technology has begun to be introduced into the physical layer of wireless communication, such as AI-based beam management and AI-based channel state information (CSI) feedback enhancement.

[0004] If a network device requires the cooperation of a terminal device to implement a corresponding AI function, the network device can query the UE's capabilities. Therefore, the terminal device needs to report to the network device information about its supported AI capabilities, such as the capabilities it can provide for the AI ​​function. Summary of the Invention

[0005] This application provides a communication method and communication system for reporting the ability of terminal devices to support AI functions related to generating CSI reports.

[0006] To achieve the above objectives, the embodiments of this application adopt the following technical solutions:

[0007] Firstly, this application provides a communication method applied to a terminal device. The terminal device acquires first CSI capability information supported by the terminal device. This first CSI capability information includes the size of the NPU and the size of the first CPU. The first CPU includes determining the CPU used by the first CSI report, or, alternatively, determining the CPU used by the first CSI report and the second CSI report. The first CSI report represents an AI-based CSI report, and the second CSI report represents a non-AI-based CSI report, i.e., a traditional CSI report.

[0008] Then, the terminal device sends the first CSI capability information to the network device.

[0009] In this application, the terminal device determines its supported AI capabilities and obtains first CSI capability information. This first CSI capability information includes the processing unit size supported by the terminal device (including the size of the NPU and the first CPU). Subsequently, the terminal device reports the first CSI capability information to the network device, thereby reporting the AI ​​capabilities supported by the terminal device related to generating the CSI report.

[0010] In one possible design, the aforementioned NPU size represents the number of NPUs shared by the AI ​​function groups supported by the terminal device. Based on this, the AI ​​function groups of the terminal device can share NPUs, thereby the terminal device reporting the size of the shared NPUs.

[0011] In one possible design approach, the NPU size includes the number of NPUs corresponding to each AI function group supported by the terminal device. Based on this, the NPUs occupied by the AI ​​function groups of the terminal device are independent, thus the terminal device reports the NPU size corresponding to each AI function group.

[0012] The aforementioned AI function group includes at least one AI function.

[0013] In one possible design, the aforementioned first CPU includes a CPU for determining the first CSI report and the second CSI report; that is, the CPU required for determining the first CSI report and the second CSI report is shared. Accordingly, the size of the first CPU represents the number of CPUs shared by the AI ​​functions supported by the terminal device and the number of CPUs used to determine the second CSI report, i.e., the number of CPUs provided by the terminal device.

[0014] In one possible design, the aforementioned first CPU includes the CPU used to determine the first CSI report; that is, the CPU required for the first CSI report and the second CSI report are independent. The first CPU is used only to determine the first CSI report. Accordingly, the size of the first CPU includes the number of CPUs corresponding to each AI function supported by the terminal device.

[0015] In one possible design, the aforementioned first CSI capability information may further include AI processing latency information supported by the terminal device. This AI processing latency information may include processing latency information corresponding to each individual AI function, processing latency information corresponding to each AI function group, or processing latency information corresponding to all AI functions. An AI function group includes at least one AI function. Based on this, the terminal device may also report AI processing latency information, which represents the additional latency caused by the AI ​​functions provided by the terminal device, thereby enabling the reporting of the AI ​​processing capabilities supported by the terminal device.

[0016] In one possible design approach, the AI ​​processing latency information indicates the latency level or latency value. In other words, the reported AI processing latency information can be represented by a latency level or a specific latency value, ensuring flexibility.

[0017] In one possible design approach, the aforementioned first CSI capability information may further include AI function groups supported by the terminal device, where each AI function group includes at least one AI function. Based on this, the reporting of the AI ​​functions supported by the terminal device is achieved.

[0018] In one possible design, the terminal device can determine M CSI reports from N CSI reports for updating; where N is a positive integer, 0≤M≤N, and M is an integer; the number of NPUs occupied by the M CSI reports is less than or equal to the number of unoccupied NPUs, and the number of unoccupied NPUs is determined based on the size of the NPUs and the number of occupied NPUs; the M CSI reports include a first CSI report and / or a second CSI report;

[0019] Subsequently, the terminal device sends N CSI reports to the network device, and among the N CSI reports, M are updated CSI reports.

[0020] Based on this, the updated CSI report can be determined by the size of the unused processing units.

[0021] Optionally, N is a positive integer greater than 1.

[0022] In one possible design, the size of the NPU represents the number of NPUs shared by the AI ​​function groups supported by the terminal device, and the size of the first CPU represents the number of CPUs shared by the AI ​​functions supported by the terminal device and the CPUs used to determine the second CSI report.

[0023] N CSI reports include N AI The first CSI report, where 0≤N AI ≤N, and N AIIt is an integer. The number of NPUs used by the M1 first CSI reports out of the M CSI reports is less than or equal to the number of unused NPUs, and the number of CPUs used by the M CSI reports is less than or equal to the number of unused CPUs.

[0024] Where 0≤M1≤N AI M1 is an integer.

[0025] Based on this, the updated CSI report is determined when both the NPU and CPU are shared between the first CSI report and the second CSI report.

[0026] Optionally, when both NPU and CPU are shared between the first CSI report and the second CSI report, the number of NPUs occupied by M1 first CSI reports is greater than the number of NPUs occupied by P1 first CSI reports, where P1 first CSI reports include N... AI At least one of the first CSI reports, P1 first CSI report is different from M1 first CSI report;

[0027] Furthermore, the number of CPUs used by M CSI reports is greater than the number of CPUs used by P2 CSI reports, and P2 CSI reports include at least one of N CSI reports. P2 CSI reports are different from M CSI reports.

[0028] Based on this, the maximum utilization of unused processing units is achieved, ensuring full utilization of resources.

[0029] Optionally, the M CSI reports have a higher priority than the other CSI reports among the N CSI reports. Based on this, the updates of the higher-priority CSI reports are implemented.

[0030] In one possible design approach, the size of the NPU includes the number of NPUs corresponding to each AI function group supported by the terminal device, and the size of the first CPU represents the number of all AI functions supported by the terminal device and the number of CPUs shared for determining the second CSI report.

[0031] N CSI reports include N AI The first CSI report, N AI The first CSI report includes K AI functional groups, each corresponding to N. k The first CSI report; where 0≤N AI ≤N, N AI It is an integer, 1≤k≤K, where k is an integer;

[0032] M CSI reports include M2 ​​AI function groups, each corresponding to Q. jThe first CSI report states that M2 AI functional groups belong to K AI functional groups, where 1 ≤ j ≤ M2, and j is an integer.

[0033] Q corresponding to the j-th AI function group j The number of NPUs used by the first CSI report is less than or equal to the number of unused NPUs corresponding to the j-th AI function group;

[0034] The number of CPUs used by M CSI reports is less than or equal to the number of CPUs not used.

[0035] Based on this, CPU sharing is achieved between the first CSI report and the second CSI report, but the updated CSI report is determined in the case of NPU independence.

[0036] Optionally, in the case of CPU sharing between the first and second CSI reports, but NPU independence, the Q corresponding to the j-th AI function group... j The number of NPUs used by the first CSI report is greater than the number of NPUs used by P3 first CSI reports. P3 first CSI reports include the NPUs corresponding to the j-th AI function group. j At least one of the first CSI reports, P3 first CSI reports and Q j The first CSI report is different;

[0037] Furthermore, the number of CPUs used by M CSI reports is greater than the number of CPUs used by P4 CSI reports. P4 CSI reports include at least one of N CSI reports, and P4 CSI reports are different from M CSI reports.

[0038] Based on this, the maximum utilization of unused processing units is achieved, ensuring full utilization of resources.

[0039] In one possible design approach, the size of the NPU includes the number of NPUs corresponding to each AI function group supported by the terminal device, and the size of the first CPU includes the number of CPUs corresponding to each AI function; the N CSI reports include the N corresponding to K AI function groups. k The first CSI report; where 1 ≤ k ≤ K, and k is an integer;

[0040] From N CSI reports, identify M CSI reports to update, including:

[0041] For each of the K AI function groups, start from the N corresponding to the AI ​​function group. k The first CSI report identified M k The first CSI report has been updated;

[0042] M corresponding to the AI ​​function group k The number of NPUs used by each first CSI report is less than or equal to the number of unused NPUs corresponding to the AI ​​function group; the number of unused NPUs corresponding to the AI ​​function group is determined based on the number of NPUs corresponding to the AI ​​function group and the number of NPUs already used.

[0043] And, the M corresponding to the AI ​​function group k The number of CPUs used in a first CSI report is less than or equal to the number of unused CPUs corresponding to the AI ​​function group; the number of unused CPUs corresponding to the AI ​​function group is determined based on the number of CPUs corresponding to the AI ​​function group and the number of CPUs already used.

[0044] Based on this, the determination of the updated CSI report is achieved when both the CPU and NPU are independent between the first CSI report and the second CSI report.

[0045] Optionally, when both the CPU and NPU are independent between the first and second CSI reports, the M corresponding to the AI ​​function group... k The number of NPUs used by one first CSI report is greater than the number of NPUs used by P5 first CSI reports. P5 first CSI reports include the NPUs corresponding to the AI ​​function group. k At least one of the first CSI reports, P5 first CSI reports and M k The first CSI report is different;

[0046] And, the M corresponding to the AI ​​function group k The CPU usage of one first CSI report is greater than the CPU usage of P6 first CSI reports, where P6 first CSI reports include N corresponding to the AI ​​function group. k At least one of the first CSI reports, P6 first CSI reports with M k The first CSI report is different.

[0047] Based on this, the maximum utilization of unused processing units is achieved, ensuring full utilization of resources.

[0048] In one possible design approach, the CSI reference resource corresponding to the first CSI report is determined based on the first CSI reference resource and AI processing latency information; wherein, the first CSI reference resource is a traditional CSI reference resource.

[0049] Based on this, the determination of CSI reference resources under the AI-based CSI report (i.e., the aforementioned first CSI report) is realized.

[0050] In one possible design approach, when the time domain characteristic of the first CSI report is aperiodic, the first latency corresponding to the first CSI report is determined based on the traditional first latency and AI processing latency information, and the second latency corresponding to the first CSI report is determined based on the traditional second latency and AI processing latency information.

[0051] The first delay (or first delay requirement) includes the minimum time required for the terminal device to decode the physical downlink control channel (PDCCH) that triggers the first CSI report and to perform measurement calculations on the measurement reference signal (or reference signal).

[0052] The second delay (or second delay requirement) includes the minimum time required for the terminal device to perform measurement calculations on the measurement reference signal.

[0053] Based on this, the first latency requirement and the second latency requirement under the non-periodic first CSI report are determined.

[0054] Secondly, this application provides a communication method applied to a network device. The network device receives first Channel State Information (CSI) capability information sent by a terminal device; wherein, the first CSI capability information includes the size of a Neural Processing Unit (NPU) and the size of a first CSI Processing Unit (CPU); the first CPU includes determining the CPU occupied by a first CSI report, or the first CPU includes determining the CPU occupied by a first CSI report and a second CSI report; the first CSI report represents a CSI report based on artificial intelligence (AI) functions, and the second CSI report represents a CSI report not based on AI functions.

[0055] In one possible design, the network device receives N CSI reports sent by the terminal device, of which M are updated CSI reports; where N is a positive integer, 0≤M≤N, and M is an integer; the number of NPUs occupied by the M CSI reports is less than or equal to the number of unoccupied NPUs, and the number of unoccupied NPUs is determined based on the size of the NPUs and the number of occupied NPUs; the M CSI reports include a first CSI report and / or a second CSI report.

[0056] Thirdly, this application provides a communication device, including a module for performing the communication method of the first aspect above; and / or a module for performing the communication method of the second aspect above.

[0057] Among them, the communication device can be used as a network device or a terminal device.

[0058] Fourthly, embodiments of this application provide a communication system that may include a terminal device and a network device. The terminal device is used to perform the communication method as described in the first aspect above, and / or the network device is used to perform the communication method as described in the second aspect above.

[0059] The terminal device acquires the first Channel State Information (CSI) capability information of the terminal device; wherein the first CSI capability information includes the size of the Neural Processing Unit (NPU) and the size of the first CSI Processing Unit (CPU); the first CPU includes determining the CPU occupied by the first CSI report, or the first CPU includes determining the CPU occupied by the first CSI report and the second CSI report; the first CSI report represents a CSI report based on artificial intelligence functions, and the second CSI report represents a CSI report not based on AI functions.

[0060] The terminal device sends the first CSI capability information to the network device.

[0061] The network device receives the first CSI capability information.

[0062] Fifthly, this application provides a communication device, comprising: at least one processor and an interface circuit, wherein the interface circuit is configured to receive signals from other communication devices outside the communication device and transmit them to the processor or to send signals from the processor to other communication devices outside the communication device, and the processor is configured to implement the communication methods of the first aspect and / or the second aspect above through logic circuits or execution code instructions.

[0063] Sixthly, this application provides a computer-readable storage medium including a computer program or instructions that, when executed on a computer, cause the computer to perform the communication methods described in the first and / or second aspects above.

[0064] In a seventh aspect, this application provides a chip, including: an interface circuit and a logic circuit. The interface circuit is used to receive signals from other chips outside the chip and transmit them to the logic circuit, or to send signals from the logic circuit to other chips outside the chip. The logic circuit is used to implement the communication methods of the first aspect and / or the second aspect above.

[0065] Eighthly, this application provides a computer program product comprising: a computer program or instructions that, when executed on a computer, cause the computer to perform the communication methods as described in the first and / or second aspects above.

[0066] It is understood that any of the communication devices, communication systems, chips, computer-readable storage media, or computer program products provided above can be applied to the corresponding methods provided above. Therefore, the beneficial effects they can achieve can be referred to the beneficial effects described in the corresponding methods. Furthermore, the communication method provided in the second aspect can also refer to the relevant descriptions and beneficial effects in the communication method provided in the first aspect, and will not be repeated here. Attached Figure Description

[0067] Figure 1 is a schematic diagram of a communication system provided in an embodiment of this application;

[0068] Figure 2A is a schematic diagram of the occupancy time of a CSI processing unit provided in an embodiment of this application;

[0069] Figure 2B is a schematic diagram of the occupancy time of a CSI processing unit provided in an embodiment of this application;

[0070] Figure 2C is a schematic diagram of the occupancy time of a CSI processing unit provided in an embodiment of this application;

[0071] Figure 3A is a schematic diagram of a latency requirement provided by an embodiment of this application;

[0072] Figure 3B is a schematic diagram of a CSI reference resource provided in an embodiment of this application;

[0073] Figure 4A is a schematic diagram of an AI function provided in an embodiment of this application;

[0074] Figure 4B is a schematic diagram of an AI function provided in an embodiment of this application;

[0075] Figure 5A is a schematic diagram of an AI function provided in an embodiment of this application;

[0076] Figure 5B is a schematic diagram of an AI function provided in an embodiment of this application;

[0077] Figure 6 is a flowchart illustrating a communication method provided in an embodiment of this application;

[0078] Figure 7A is a schematic diagram of a processing unit provided in an embodiment of this application;

[0079] Figure 7B is a schematic diagram of a processing unit provided in an embodiment of this application;

[0080] Figure 7C is a schematic diagram of a processing unit provided in an embodiment of this application;

[0081] Figure 8 is a schematic diagram of a CSI reference resource provided in an embodiment of this application. Detailed Implementation

[0082] In the description of this application, unless otherwise stated, " / " indicates that the objects before and after are in an "or" relationship. For example, A / B can mean A or B. "And / or" in this application is merely a description of the relationship between the related objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist simultaneously, and B exists alone. A and B can be singular or plural.

[0083] In the description of this application, unless otherwise stated, "multiple" means two or more. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of a single item or a plurality of items. For example, at least one of a, b, or c can mean: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or multiple.

[0084] Furthermore, to facilitate a clear description of the technical solutions in the embodiments of this application, the terms "first," "second," "1," "2," "A," "B," and "C" are used in the embodiments of this application to distinguish identical or similar items with essentially the same function and effect. Those skilled in the art will understand that the terms "first," "second," etc., do not limit the quantity or execution order, and that the terms "first," "second," etc., are not necessarily different.

[0085] In the embodiments of this application, the terms "exemplary" or "for example" are used to indicate that something is an example, illustration, or description. Any embodiment or design that is described as "exemplary" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design. Specifically, the use of terms such as "exemplary" or "for example" is intended to present the relevant concepts in a specific manner to facilitate understanding.

[0086] It is understood that the term "embodiment" used throughout the specification means that a specific feature, structure, or characteristic related to an embodiment is included in at least one embodiment of this application. Therefore, various embodiments throughout the specification do not necessarily refer to the same embodiment. Furthermore, these specific features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. It is understood that in the various embodiments of this application, the sequence number of each process does not imply the order of execution; the execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0087] It is understood that in this application, "...when" and "if" both refer to the corresponding processing that will be carried out under certain objective circumstances, and are not limited to a specific time, nor do they require a judgment action to be performed during implementation, nor do they imply any other limitations.

[0088] It is understood that some optional features in the embodiments of this application can be implemented independently in certain scenarios without relying on other features, such as the current solution on which they are based, to solve the corresponding technical problems and achieve the corresponding effects. Alternatively, they can be combined with other features as needed in certain scenarios. Correspondingly, the apparatus given in the embodiments of this application can also implement these features or functions, which will not be elaborated here.

[0089] In this application, unless otherwise specified, the same or similar parts between the various embodiments can be referred to each other. Unless otherwise specified or logically conflicting, the terminology and / or descriptions between different embodiments are consistent and can be mutually referenced. Different embodiments can be combined to form new embodiments based on their inherent logical relationships. The following descriptions of the embodiments of this application do not constitute a limitation on the scope of protection of this application.

[0090] This application provides a communication system, as shown in Figure 1, which includes network equipment and terminal equipment. A network device is a means deployed in a radio access network to provide wireless communication functions for terminal equipment. Network equipment can include various forms of macro base stations, micro base stations (also called small stations), relay stations, access points, etc. In systems employing different radio access technologies, the name of the network device may differ, such as a base transceiver station (BTS) in a Global System for Mobile Communication (GSM) or Code Division Multiple Access (CDMA) network, a node B (NB) in Wideband Code Division Multiple Access (WCDMA), and an evolved Node B (eNB) in Long Term Evolution (LTE). The network device can also be a radio controller in a cloud radio access network (CRAN) scenario. Network equipment can also be base station equipment in fifth-generation (5G) mobile communication systems or next-generation wireless communication networks, or network equipment in future evolved Public Land Mobile Network (PLMN) networks. Network equipment can also be wearable devices or vehicle-mounted devices. Network equipment can also be transmission and reception points (TRPs).

[0091] The aforementioned terminal devices can be environmental IoT devices, including various handheld devices, in-vehicle devices, wearable devices, computing devices, or other processing devices connected to a wireless modem with wireless communication capabilities. Terminals can be mobile stations (MS), subscriber units, cellular phones, smartphones, wireless data cards, personal digital assistant (PDA) computers, tablet computers, wireless modems, handsets, laptop computers, machine-type communication (MTC) terminals, etc.

[0092] The aforementioned communication systems can be applied to Long Term Evolution (LTE) systems, Universal Mobile Telecommunications System (UMTS) systems, Code Division Multiple Access (CDMA) systems, Wireless Local Area Network (WLAN) systems, or the Fifth Generation (5G) systems or next-generation wireless communication systems, etc.

[0093] For ease of understanding, this application will first introduce the basic technologies involved in this application.

[0094] 1. CSI processing unit (CPU)

[0095] Network devices send reference signals to terminal devices, such as channel state information reference signals (CSI-RS) and synchronization signal blocks (SSBs). Terminal devices receive these reference signals, perform measurements and calculations based on them, and obtain the corresponding measurement results, i.e., a CSI report. Afterward, the terminal device sends the CSI report back to the network device. The resources required for the terminal device to measure and calculate the reference signals are called the CSI processing unit.

[0096] 2. CPU usage

[0097] If the terminal device supports N at the same time CPU If N CSI calculations are performed, then the terminal device can be said to have N CPU There are L CPUs used for processing CSI reports. On a given orthogonal frequency division multiplexing (OFDM) symbol, if L CPUs are already occupied, the terminal device has N CPUs available. CPU -L unused CPUs. On the same OFDM symbol (which can be understood as at the same time), the terminal device performs corresponding CSI measurements on N CSI reports. A CSI measurement represents a measurement calculation of the reference signal, where N is a positive integer. The N CSI reports correspond to N unused CPUs. CPU -L unused CPUs are now being used. The number of CSI processing units used by each CSI report is as follows: The terminal device can sort N CSI reports in descending order of priority and select the top M CSI reports for corresponding CSI measurements, where 0 ≤ M ≤ N. Then, the total number of CSI processing units used by these M CSI reports satisfies... in, This represents the CPU usage (or CPU size) of the nth CSI report out of N CSI reports. n = 0, ..., N-1, where a larger n indicates a higher priority.

[0098] In simple terms, when there are not enough CPUs, the terminal device discards low-priority CSI reports and updates the M highest-priority CSI reports.

[0099] 3. CPU usage time

[0100] For CSI reports where the reported value is not null (none), the terminal device can determine the CPU usage time based on the time-domain characteristics (or time-domain features, report configuration type) of the CSI report. These time-domain characteristics include periodicity, semi-persistentity, and aperiodicity.

[0101] For CSI reports with periodic or semi-persistent time-domain characteristics (excluding the first semi-persistent CSI report triggered by the physical downlink control channel (PDCCH) and carried by the physical uplink shared channel (PUSCH), as shown in Figure 2A, the CSI processing unit's occupancy time begins from the first OFDM symbol of the most recent reference signal resource set in the reference signal resource set used for channel or interference measurement, and ends with the last OFDM symbol of the PUSCH / PUCCH carrying the CSI report. Here, the most recent reference signal resource set is the last reference signal resource set in the reference signal resource set no later than the CSI reference resource corresponding to the CSI report. Furthermore, this reference signal resource set can also be referred to as the CSI resource set, such as CSI-RS, CSI-IM, SSB, etc. Alternatively, the reference signal resource set can be simply described as CSI resources.

[0102] For the first semi-persistent CSI report carried by PUSCH triggered by PDCCH, see Figure 2B. The CSI processing unit occupancy time starts from the first OFDM symbol after PDCCH and ends at the last OFDM symbol of the PUSCH carrying the CSI report.

[0103] For CSI reports with a non-periodic time-domain characteristic, see Figure 2C. The CSI processing unit occupancy time starts from the first OFDM symbol used to trigger the CSI report, i.e., the first OFDM symbol after the PDCCH, and ends at the last OFDM symbol of the PUSCH carrying the CSI report.

[0104] 4. Traditional CSI calculation time

[0105] After the terminal device receives the reference signal (i.e., the aforementioned reference signal resource set) sent by the network device, the terminal device needs a certain amount of time to measure and calculate the reference signal, that is, to perform CSI measurement and calculation.

[0106] For CSI reports with a non-periodic time-domain characteristic, if the terminal device determines that the first and second conditions are met, indicating that the terminal device has sufficient time to measure and calculate CSI, then the terminal device can generate a valid non-periodic CSI report.

[0107] In this context, both the first and second conditions are time-related conditions. The first condition indicates that the time from the end of the first channel to the beginning of the second channel must be greater than or equal to the first delay requirement Z. The first channel is the downlink channel (such as PDCCH) used to trigger the aperiodic CSI report. The second channel is the uplink channel (such as PUSCH) carrying the aperiodic CSI report. Referring to Figure 3A, the first condition represents the time delay T corresponding to the time from the end of the PDCCH to the beginning of the PUSCH carrying the aperiodic CSI report. proc,CSI Greater than or equal to Z.

[0108] In this embodiment of the application, the first delay requirement is also referred to as the traditional first delay requirement, or the traditional first delay, or the conventional first delay.

[0109] The second condition indicates that the time from the end of the first reference signal to the beginning of the second channel must be greater than or equal to the second delay requirement Z'. The first reference signal refers to the last reference signal used for channel measurement in the aperiodic CSI report (i.e., the aforementioned set of most recent reference signal resources). Referring to Figure 3A above, the second condition indicates the time delay T corresponding to the time from the end of the most recent reference signal resource set to the beginning of the PUSCH carrying the aperiodic CSI report. p ' roc,CSI Greater than or equal to Z'.

[0110] In this embodiment of the application, the second delay requirement is also referred to as the traditional second delay requirement, or the traditional second delay, or the conventional second delay.

[0111] Specifically, the above Where M represents the M beams that need to be updated and reported, Z(m) represents the traditional first delay of the m-th aperiodic CSI report, and Z′(m) represents the traditional second delay of the m-th aperiodic CSI report. In other words, if the first channel only triggers the m-th CSI report, then the traditional first delay Z = Z(m), and the traditional second delay Z′ = Z(m). However, the first channel may trigger multiple CSI reports simultaneously, so the maximum value should be selected when calculating Z and Z′.

[0112] In the protocol, the values ​​of Z(m) and Z′(m) correspond to preset values, and their specific values ​​are related to factors such as the amount of CSI report submitted and the subcarrier spacing. For example, when the reported amount is cri-RSRP, (Z(m), Z′(m)) = (Z3, Z′3), and (Z3, Z′3) corresponds to a predefined table in the protocol, as shown in Table 1. When the subcarrier spacing parameter μ = 0, Z3 = 22, Z′3 = X. μ＝0 , where X μThe specific value is related to the reporting capability of the terminal device. Of course, (Z(m), Z′(m)) may also be equal to (Z1, Z′1) shown in Table 1, or (Z2, Z′2).

[0113] Table 1

[0114] It should be noted that the CSI report mentioned above is a conventional CSI report, i.e., a traditional CSI report, which is not based on AI functionality. The aforementioned conventional first latency can be understood as including at least the minimum time required for the terminal device to decode the PDCCH that triggers the conventional CSI report and to perform measurement calculations on the reference signal used to determine the conventional CSI report. Optionally, this time is in OFDM symbols.

[0115] Similarly, the aforementioned conventional second delay can be understood as including at least the minimum time required for the terminal equipment to perform measurement calculations on the reference signal used to determine the conventional CSI report. Optionally, this time is also in OFDM symbols, meaning the conventional second delay includes the number of OFDM symbols required to calculate beam measurement information.

[0116] Of course, traditional first-order latency may also include other delays, such as redundant latency. Similarly, traditional second-order latency may also include other delays, such as redundant latency.

[0117] In general, conventional first latency can be represented as the minimum time required for a terminal device to move from the end of the PDCCH that triggers the non-periodic conventional CSI report to the beginning of the PUSCH that carries the conventional CSI report (i.e., from the first symbol after the PDCCH that triggers the conventional CSI report to the beginning of the first uplink symbol of the PUSCH that carries the conventional CSI report).

[0118] The conventional second delay can be represented as the minimum time required from the end of the aforementioned first reference signal (or conventional first reference signal) to the beginning of the PUSCH carrying the conventional CSI report.

[0119] For CSI reports with periodic or semi-persistent time-domain characteristics, since there is no concept of CSI calculation time, but as mentioned above, the reference signal used for CSI measurement in periodic or semi-persistent CSI reports needs to be no later than the CSI reference resource. In the time domain, this CSI reference resource (here referred to as the traditional CSI reference resource) is defined as follows: As shown in Figure 3B, if a CSI report is reported in uplink time slot n′, then the CSI reference resource can be defined as a downlink time slot, i.e., nn. CSI_ref , where n is the downlink time slot corresponding to uplink time slot n′, and n can be calculated from n′. CSI_ref This indicates the time slot offset.

[0120] The above section introduced the relevant technologies under the traditional CSI framework. The following section will introduce the relevant technical solutions under the AI-based CSI framework involved in this application.

[0121] With low-frequency communication resources becoming increasingly scarce, wireless communication has begun to evolve towards higher-frequency bands. However, the high attenuation, low penetration, and easy absorption characteristics of high-frequency electromagnetic waves necessitate larger antennas for terminal devices and other communication equipment to ensure communication quality. While large-scale antennas can improve signal strength, they also incur significant overhead. Examples include CSI feedback overhead in beam management and channel state information acquisition, and downlink reference signal overhead on the network side. To reduce this overhead, AI technology is being introduced into the physical layer of wireless communication.

[0122] At the physical layer, AI functions (or AI-based use cases) mainly include one or more of the following: AI-based beam management, AI-based CSI feedback enhancement, and AI-based positioning enhancement.

[0123] The aforementioned AI-based beam management mainly includes the following two sub-functions (or sub-use cases):

[0124] Use Case 1-1, Spatial Domain Beam Prediction: A network device sends a small first set of reference signals, Set 1, to a terminal device. The terminal device then measures and calculates the reference signals in Set 1. Based on the measurement results of Set 1, the terminal device's AI model predicts the signal quality of a larger second set of reference signals, Set 2. That is, as shown in Figure 4A, the terminal device uses the measured reference signal receiving power (RSRP) values ​​of the beams in Set 1, along with possible other auxiliary information (such as beam index information), as input parameters to the relevant AI model. The AI ​​model then outputs the measured reference signal receiving power values ​​for each beam in Set 2 and / or the probability that each beam is the optimal beam.

[0125] Use Case 1-2, Timing Beam Prediction: The network device sends a first reference signal set Set 1 to the terminal device. The terminal device measures the reference signals in Set 1 at historical time T1, obtaining historical measurement results for historical time T1. Based on these historical measurement results, it predicts the signal quality of a second reference signal set Set 2 at future time T2. In other words, as shown in Figure 4B, the terminal device uses the measured reference signal received power values ​​of the beams in Set 1 at historical time T1, along with other possible auxiliary information, as input parameters to a relevant AI model. The AI ​​model then outputs the measured reference signal received power values ​​of each beam in Set 2 at future time T2 and / or the probability that each beam is the optimal beam.

[0126] The aforementioned AI-based CSI feedback enhancement mainly includes the following two sub-use cases:

[0127] Use Case 2-1: Spatial-Frequency Domain CSI Compression Based on a Dual-End AI Model: This model consists of a CSI generation part model on the terminal device side and a CSI reconstruction part model on the network device side. As shown in Figure 5A, the terminal device uses the measurement results of the reference signal as input parameters to the CSI generation part model, enabling it to perform AI inference, generate, and output a compressed CSI. The terminal device then feeds this compressed CSI back to the network device. The network device uses this compressed CSI as input parameters to the CSI reconstruction part model, enabling it to reconstruct the CSI based on the compressed CSI, generate, and output a reconstructed CSI.

[0128] Use Case 2-2: CSI Prediction Based on AI Model on Terminal Device Side: As shown in Figure 5B, the terminal device uses historical CSI as the input parameter of the CSI prediction model to predict the CSI at one or more future moments.

[0129] Since the aforementioned AI positioning enhancement is not a real-time requirement and does not involve the aforementioned CPU usage, it has little relevance to the technical solution provided in this application. Therefore, this application will not describe it.

[0130] Under the AI-based CSI framework, the terminal device needs to report to the network device its supported AI capabilities, such as the capabilities it can provide for AI functions. Referring to the previous introduction to AI functions, AI functions also require measurement and calculation of reference signals. That is, similar to traditional CSI reporting, AI-based CSI reporting also requires CPU utilization. Therefore, the terminal device's capabilities need to consider whether the CPU counts are shared or separate between AI-based CSI reports (or the first CSI report) and traditional CSI reports (or the second CSI report), and whether the network processing unit (NPU) counts are shared or separate between AI functions. The NPU is similar to the CPU mentioned earlier, except that the NPU is used for the AI ​​model. Furthermore, the NPU in this application can be replaced with an AI processing unit (APU).

[0131] The process of a terminal device reporting AI capability information will be described in detail below, with reference to Figure 6. As shown in Figure 6, this process (i.e., the communication method) may include:

[0132] S201. The terminal device obtains the first CSI capability information of the terminal device. The first CSI capability information includes the size of the NPU and the size of the first CPU. The first CPU includes determining the CPU used by the first CSI report, or, the first CPU includes determining the CPU used by the first CSI report and the second CSI report.

[0133] Wherein, the aforementioned first CSI capability information represents the capability information regarding the AI ​​function. The aforementioned NPU size represents the number of NPUs, and the first CPU size represents the number of first CPUs. The first CPU represents the CPU associated with determining the first CSI report. Optionally, the first CPU is used to determine the first CSI report. Alternatively, the first CPU is used to determine the first CSI report and the second CSI report.

[0134] In some embodiments, as shown in Figure 7A, the NPU is shared among AI function groups. That is, the AI ​​functions share the NPU, and an AI function group is a collection of AI functions that includes at least one AI function. For example, all the AI ​​functions used in use cases 1-1, 1-2, 2-1, and 2-2 above constitute an AI function group. As another example, the AI ​​functions used in use cases 1-1 and 1-2 constitute an AI function group. Yet another example is that use case 2-1 constitutes an AI function group.

[0135] Accordingly, the size of the NPU mentioned above represents the number NPUs shared by the AI ​​function groups supported by the terminal device. NPU .

[0136] In other embodiments, the NPU is not shared among the AI ​​function groups. For example, the terminal device has K AI function groups, and each AI function group has a corresponding NPU (as shown in Figures 7B and 7C). Accordingly, the size of the NPU includes the number of NPUs corresponding to each AI function group supported by the terminal device, such as the number of NPUs corresponding to the k-th AI function group being...

[0137] Optionally, as shown in Figure 7A or Figure 7B above, the CPU between the first CSI report and the second CSI report can be shared, meaning that CSI signal measurement calculations for different types of CSI reports share the same CPU. Accordingly, the first CPU described above is used not only for determining the CSI signal measurements associated with the first CSI report but also for determining the CSI signal measurements associated with the second CSI report. In other words, the first CPU described above includes a CPU for determining the first CSI report and the second CSI report.

[0138] Alternatively, the CPUs for the first and second CSI reports can be independent, meaning that CSI signal measurement calculations for different types of CSI reports do not share the same CPU. For example, the terminal device has K AI function groups, each with its own corresponding CPU (as shown in Figure 7C). Accordingly, the aforementioned first CPU is only used for CSI signal measurements related to determining the first CSI report. In other words, the aforementioned first CPU includes the CPU used to determine the first CSI report. The CSI signal measurements related to determining the second CSI report are implemented through a conventional CPU (or second CPU).

[0139] In general, the first CSI capability information reported by the terminal device can include the processing unit size. In Case 1, as shown in Figure 7A above, all AI functions supported by the terminal device share the NPU, and CSI signal measurement calculations share the CPU. Accordingly, the processing unit size reported by the terminal device can include the number of CPUs N in the terminal device. CPU The number of NPUs in the terminal device N NPU .

[0140] Scenario 2, as shown in Figure 7B above, involves AI function groups supported by the terminal device each having their own corresponding NPU, but CSI signal measurement and calculation sharing the CPU. Accordingly, the size of the processing unit reported by the terminal device can include the number of CPUs N in the terminal device. CPU The number of NPUs corresponding to each of the K AI function groups supported by the terminal device.

[0141] Scenario 3, as shown in Figure 7C above, involves AI function groups supported by the terminal device, each with its own corresponding NPU and CPU. Accordingly, the processing unit size reported by the terminal device can include the number of CPUs and NPUs corresponding to each of the K AI function groups supported by the terminal device.

[0142] In some embodiments, the first CSI capability information may further include the AI ​​function groups supported by the terminal device, such as each of the K AI function groups mentioned above.

[0143] In some embodiments, the first CSI capability information may further include AI processing latency information supported by the terminal device, which includes the time consumed by the AI ​​model of the AI ​​function, such as the latency of activating / deactivating the AI ​​model corresponding to the AI ​​function, and the latency of inference through the AI ​​model corresponding to the AI ​​function.

[0144] The aforementioned AI processing latency information may include the processing latency information corresponding to each AI function. That is, each AI function has its own corresponding processing latency information, and the processing latency information between AI functions may be the same or different.

[0145] Alternatively, the aforementioned AI processing latency information may include processing latency information corresponding to each AI function group, meaning that each AI function group has its own corresponding processing latency information, and the processing latency information between AI function groups may be the same or different. Optionally, the processing latency information corresponding to each AI function within the same AI function group may also be the same or different.

[0146] Alternatively, the aforementioned AI processing latency information may include the processing latency information corresponding to each of the AI ​​function groups, that is, the terminal device reports a processing latency information for all AI function groups supported by the terminal device.

[0147] Optionally, the AI ​​processing latency information (or the aforementioned processing latency information) can indicate latency levels, with each latency level corresponding to a specific latency value. For example, the latency value corresponding to latency level i is δ. i .

[0148] In some embodiments, the AI ​​processing latency information may be in units of the aforementioned time slots or other units, such as milliseconds. Additionally, the AI ​​processing latency information may include uplink AI processing latency information and / or downlink AI processing latency information. This application does not limit these aspects.

[0149] S202, The terminal device sends the first CSI capability information to the network device.

[0150] S203. The network device receives the first CSI capability information sent by the terminal device.

[0151] In this embodiment of the application, after receiving the first CSI capability information, the network device can determine the terminal device's capabilities regarding AI functions based on the first CSI capability information, and thus configure the corresponding first CSI report (i.e., AI-CSI report).

[0152] In some embodiments, the terminal device determines M CSI reports from N CSI reports for updating. Here, N is a positive integer greater than 1, 0 ≤ M ≤ N, and M is an integer. The number of NPUs occupied by the M CSI reports is less than or equal to the number of unoccupied NPUs, which is determined based on the size of the NPUs and the number of occupied NPUs. The M CSI reports include a first CSI report and / or a second CSI report. Alternatively, the N CSI reports may also include a first CSI report and / or a second CSI report.

[0153] Subsequently, the terminal device sends N CSI reports to the network device, of which M are updated CSI reports. The network device then receives these N CSI reports.

[0154] Optionally, the determination of the aforementioned M CSI reports is related to the unit size supported by the terminal device. The determination process for the M CSI reports will be described below in conjunction with the three scenarios described above.

[0155] For case 1 (i.e., CSI measurement calculation (i.e., reference signal measurement calculation) sharing N) CPU One CPU, shared by all AI function groups supported by the terminal device. CPU (NPUs), the above N CSI reports start occupying CPUs and NPUs from the same OFDM symbol. Similar to the previous text, on this OFDM symbol, if L1 CPUs are already occupied (i.e., the number of occupied CPUs is L1), then the number of unoccupied CPUs is based on N. CPU And L1 determines the number of unused CPUs N CPU -L1. Similarly, if L2 NPUs are already occupied (i.e., the number of occupied NPUs is L2), then the number of unoccupied NPUs is determined by N. NPU And L2 determines the number of unused CPUs N NPU -L2.

[0156] The above N CSI reports include N AI The first CSI report, 0≤N AI ≤N, and N AIIt is an integer. The number of NPUs used by the M1 first CSI reports out of the M CSI reports is less than or equal to the number of unused NPUs, and the number of CPUs used by the M CSI reports is less than or equal to the number of unused CPUs. 0 ≤ M1 ≤ N AI M1 is an integer. Based on this, ensure that the number of unused NPUs and CPUs is sufficient to update M CSI reports.

[0157] Optionally, the number of NPUs occupied by the M1 first CSI reports is greater than the number of NPUs occupied by the P1 first CSI reports, where the P1 first CSI reports include N... AI At least one of the first CSI reports, P1 first CSI report is different from M1 first CSI report.

[0158] Furthermore, the number of CPUs used by the aforementioned M CSI reports is greater than the number of CPUs used by the P2 CSI reports. The P2 CSI reports include at least one of the N CSI reports, and the P2 CSI reports are different from the M CSI reports.

[0159] In this context, P1 first CSI reports differing from M1 first CSI reports indicates that P1 first CSI reports are completely different from M1 first CSI reports or are partially different. Similarly, P2 CSI reports differing from M CSI reports indicates that P2 CSI reports are completely different from M CSI reports or are partially different.

[0160] In this embodiment, the M1 first CSI reports in the M CSI reports make the maximum use of the number of unused NPUs, and the M CSI reports make the maximum use of the number of unused CPUs, thereby ensuring full utilization of resources.

[0161] Optionally, similar to the above, the terminal device can prioritize updating high-priority CSI reports. That is, the priority of the aforementioned M CSI reports is higher than the priority of the other CSI reports among the aforementioned N CSI reports, thereby ensuring that high-priority CSI reports can be updated.

[0162] The process of determining M CSI reports under scenario 1 above will be illustrated below with a specific example.

[0163] First, N AI The first CSI report started occupying unused N from the same symbol. NPU -L2 NPUs, N AI The number of NPUs used by each of the first CSI reports is n = 0, 1, ..., N AI-1, where n is sorted in descending order of priority; the smaller n is, the higher the priority. Accordingly, the terminal device does not need to update N. AI The top N priority rankings in the first CSI report AI -M AI The lowest first CSI report. M AI It makes The largest possible value of M. It is understood that, in this embodiment, the smaller the n, the higher the priority of the corresponding CSI report (such as the first CSI report mentioned above). This is merely an example; of course, it is also possible that the smaller n is, the lower the priority of the corresponding CSI report. Furthermore, the size of n may be unrelated to the priority level, and this application does not impose any limitations on it.

[0164] Afterwards, the terminal device is based on N, which does not require updates. AI -M AI From the first CSI report and N CSI reports, determine the remaining CSI reports, and from the remaining CSI reports, determine M more CSI reports. For example, the remaining N' = N - (N AI -M AI ) CSI reports begin occupying unoccupied N symbols. CPU -L1 CPUs, the number of CPUs used by each of the remaining N' CSI reports is Here, n is also sorted in descending order of priority among the remaining N' CSI reports; the smaller n is, the higher the priority. Correspondingly, the terminal device does not need to update the CSI reports with the lowest priority (N'-M2), where M2 is the priority that makes... The largest possible value of M. In other words, the terminal device updates the top M2 CSI reports with the highest priority, or, more specifically, updates the top M CSI reports out of N CSI reports. The top M2 CSI reports can include M1 first CSI reports.

[0165] In this embodiment, since the first CSI report requires both CPU and NPU, while the second CSI report only requires CPU, the terminal device can first determine whether the first CSI report does not need to be updated through the NPU, i.e., determine whether the NPU is sufficient. If it is insufficient, then there is no need to update the NPU. AI If there is a first CSI report, then the updated CSI report can be determined only from the second CSI report among the N CSI reports, thus improving the efficiency of determining the updated CSI report. Of course, the terminal device can also first determine the CSI reports that do not need to be updated through the CPU, and then determine the updated CSI report through the NPU. The process is similar to the process of determining the M CSI reports in the case of 1 above.

[0166] For scenario 2 (i.e., CSI measurement calculation shared N) CPU (Each of the K AI function groups supported by the terminal device has a corresponding NPU). The aforementioned N CSI reports occupy the CPU and NPU starting from the same OFDM symbol, and the N CSI reports include N... AI The first CSI report, N AI The first CSI report includes K AI functional groups, each corresponding to N. k The first CSI report, namely N k This represents the number of first CSI reports corresponding to the k-th AI function group. Similar to the previous example, L1 CPUs are already occupied (i.e., the number of occupied CPUs is L1), and the number of unoccupied CPUs is determined by N. CPU The number of unused CPUs, determined by L1, is N. CPU -L1. For the k-th AI function group... One NPU has been occupied (i.e., the number of NPUs already occupied is...). The number of unused NPUs is based on and The number of unused NPUs is definite.

[0167] The aforementioned M CSI reports include M2 ​​AI function groups, each corresponding to a Q. j The first CSI report indicates that M2 AI function groups belong to K AI function groups, where 1 ≤ j ≤ M2, and j is an integer. Correspondingly, the Q corresponding to the j-th AI function group... j The number of NPUs used by the first CSI report is less than or equal to the number of unused NPUs corresponding to the j-th AI function group.

[0168] The number of CPUs used by the M CSI reports is less than or equal to the number of unused CPUs. Based on this, ensure that the number of unused NPUs and CPUs is sufficient to update the M CSI reports.

[0169] Optionally, the Q corresponding to the j-th AI function group j The number of NPUs used by the first CSI report is greater than the number of NPUs used by P3 first CSI reports. P3 first CSI reports include the NPUs corresponding to the j-th AI function group. j At least one of the first CSI reports, P3 first CSI reports and Q j The first CSI report is different.

[0170] Furthermore, the number of CPUs used by M CSI reports is greater than the number of CPUs used by P4 CSI reports. P4 CSI reports include at least one of N CSI reports, and P4 CSI reports are different from M CSI reports.

[0171] Among them, the P3 first CSI report corresponding to the j-th AI function group and Q j The different representations in the first CSI report P3 and Q indicate that the first CSI report is different from the Q report. j The first CSI report is completely or partially different. Similarly, the fact that the P4 CSI reports are different from the M CSI reports also means that the P4 CSI reports are completely or partially different from the M CSI reports.

[0172] In this embodiment, the updated first CSI report corresponding to each of the M2 AI function groups makes the maximum use of the number of unused NPUs corresponding to that AI function group, and the M CSI reports make the maximum use of the number of unused CPUs, thereby ensuring full utilization of resources.

[0173] Optionally, similar to the above, the terminal device can prioritize updating high-priority CSI reports to ensure that high-priority CSI reports can be updated.

[0174] The process of determining M CSI reports under scenario 2 above will be illustrated below with a specific example.

[0175] First, the N corresponding to the k-th AI function group k The first CSI report occupies the unoccupied space corresponding to the k-th AI function group, starting from the same OFDM symbol. NPUs, N k The number of NPUs used by each of the first CSI reports is Here, n = 0, 1, ..., N k -1, where n is sorted in descending order of priority; the smaller n is, the higher the priority. Correspondingly, the terminal device does not need to update N corresponding to the k-th AI function group. k The top N priority rankings in the first CSI report k -M k The lowest first CSI report. M k It makes The largest possible value of M.

[0176] Afterwards, the terminal device is based on the principle that no updates are required. From the first CSI report and N CSI reports, determine the remaining CSI reports, and then continue to determine M CSI reports from the remaining CSI reports. For example, the number of remaining CSI reports... The remaining N' CSI reports begin occupying the unoccupied N symbols in the same symbol. CPU -L1 CPUs, the number of CPUs used by each of the remaining N' CSI reports is Here, n is also sorted in descending order of priority among the remaining N' CSI reports; the smaller n is, the higher the priority. Correspondingly, the terminal device does not need to update the CSI reports with the lowest priority (N'-M'), where M' is the priority that makes... The largest value of M that is valid. In other words, the terminal device updates the top M' highest priority CSI reports, or, in other words, the terminal device updates the top M highest CSI reports out of N CSI reports.

[0177] It is understandable that when the CSI report in the top M' CSI reports is the first CSI report, it indicates that the first CSI report belongs to a certain AI function group (such as the jth AI function group). Accordingly, the number of first CSI reports corresponding to the jth AI function group included in the top M' CSI reports can be determined.

[0178] In this embodiment of the application, similar to the previous one, the terminal device can first determine the CSI reports that do not need to be updated through the CPU, and then determine the updated CSI reports through the NPU. The process is similar to the determination process of the M CSI reports in case 2 above.

[0179] The process for determining the updated M CSI reports under scenarios 1 and 2 has been described above. The process for determining the M CSI reports under scenario 3 will be described below.

[0180] For scenario 3 (i.e., each of the K AI function groups supported by the terminal device has a corresponding NPU and CPU, and the CPU size corresponding to the kth AI function group is...), NPU size is The aforementioned N CSI reports include N corresponding to each of the K AI function groups. k The first CSI report. For the k-th AI function group, if N k The first CSI report occupies CPU and NPU starting from the same OFDM symbol. Similarly, as mentioned earlier, the k-th AI function group corresponds to... One CPU has been occupied (i.e., the number of CPUs already occupied is...). The number of unused CPUs corresponding to the k-th AI function group is based on... and The number of unused CPUs is definite. Similarly, One NPU is already occupied. The number of unoccupied NPUs corresponding to the k-th AI function group is based on... and The number of unused NPUs is definite.

[0181] Accordingly, the aforementioned M CSI reports actually include the first CSI report corresponding to each of the K AI function groups supported by the terminal device. Therefore, the process of determining the aforementioned M CSI reports may include: for each of the K AI function groups, the terminal device retrieves the N reports corresponding to that AI function group... k The first CSI report identified M k The first CSI report was updated.

[0182] The M corresponding to this AI function group k The number of NPUs used by each first CSI report is less than or equal to the number of unused NPUs corresponding to that AI function group. And, the M corresponding to that AI function group... k The number of CPUs used by the first CSI report is less than or equal to the number of unused CPUs corresponding to that AI function group. Based on this, ensure that the number of unused NPUs and CPUs corresponding to the k-th AI function group is sufficient to update the M corresponding to that AI function group. k The first CSI report.

[0183] Optionally, the M corresponding to the k-th AI function group k The number of NPUs used by the kth first CSI report is greater than the number of NPUs used by P5 first CSI reports. P5 first CSI reports include the NPUs corresponding to the kth AI function group. k At least one of the first CSI reports, P5 first CSI reports and M k The first CSI report is different.

[0184] And, the M corresponding to the kth AI function group k The CPU usage of one first CSI report is greater than the CPU usage of P6 first CSI reports, where P6 first CSI reports include N corresponding to the k-th AI function group. k At least one of the first CSI reports, P6 first CSI reports with M k The first CSI report is different.

[0185] The above P5 first CSI reports and M k The different representations of the first CSI report P5 and M k All of the first CSI reports are different from each other, that is, each of the five first CSI reports is different from M.k All the first CSI reports in the first CSI report are different. Or, the above P5 first CSI reports are different from M. k The different representations of the first CSI report P5 and M k The first CSI report is different from the first CSI report in section P5, that is, the first CSI report in section P5 is different from the first CSI report in section M. k There are identical first CSI reports.

[0186] Similarly, the aforementioned P6 first CSI reports and M k The different representations of the six first CSI reports compared to M indicate that P6 first CSI reports are different. k The first CSI report is completely or partially different.

[0187] In this embodiment, the updated first CSI report corresponding to the AI ​​function group makes the most of the number of unused NPUs and CPUs corresponding to the AI ​​function group, thereby ensuring full utilization of resources.

[0188] Optionally, similar to the above, the terminal device can prioritize updating high-priority CSI reports to ensure that high-priority CSI reports can be updated.

[0189] The following example illustrates how to determine the updated M corresponding to the k-th AI function group under scenario 3. k The process of the first CSI report.

[0190] For the k-th AI function group, the terminal device does not need to update the N corresponding to the k-th AI function group. k The top N priority rankings in the first CSI report k -M k The lowest first CSI report, i.e., updated N k The top priority ranking in the first CSI report M k The highest first CSI report. M k =min(M1, M2), where M1 is the value that makes... The largest possible value of M, M², is the value that makes M true. The largest possible value of M.

[0191] The above section introduced the processing unit occupancy mechanism under the AI-enabled CSI framework. Similar to the traditional CSI framework, the AI-enabled CSI framework also involves CSI reference resources. The following section will detail the CSI reference resources under the AI-enabled CSI framework.

[0192] The CSI reference resource corresponding to the aforementioned first CSI report is determined based on the first CSI reference resource and AI processing latency information. This first CSI reference resource is the traditional CSI reference resource under the traditional CSI framework described above. For example, the CSI reference resource corresponding to the first CSI report is equal to the first CSI reference resource minus the processing latency information corresponding to the AI ​​function in the first CSI report.

[0193] Specifically, the processing latency information corresponding to the first CSI report is a specific latency value in units of downlink time slots, obtained through... This indicates that, correspondingly, as shown in Figure 8, in the time domain, the CSI reference resource corresponding to the first CSI report is equal to the traditional CSI reference resource (i.e., nn). CSI_ref )

[0194] Optionally, the CSI reference resource corresponding to the first CSI report can be used to determine the occupancy time of the CSI processing unit (see Figure 8 above). It is understood that the CPU and NPU occupancy times for any AI function are the same. Of course, the CSI reference resource corresponding to the first CSI report can also be applied to other scenarios, and this application does not limit its application. Additionally, optionally, the temporal characteristics of the first CSI report can be periodic or semi-continuous.

[0195] In some embodiments, similar to the traditional CSI framework, the AI-based CSI framework also involves a first latency. The first latency corresponding to the first CSI report is determined based on the traditional first latency under the traditional CSI framework described above and the AI ​​processing latency information introduced above. The second latency corresponding to the first CSI report is determined based on the traditional second latency and the AI ​​processing latency information.

[0196] The meaning of the first delay corresponding to the first CSI report is similar to the traditional first delay described above. For example, this first delay at least includes the minimum time required for the terminal device to decode the PDCCH that triggers the first CSI report and to perform measurement calculations on the measurement reference signal. Of course, this first delay may also include other delays. Specifically, the first delay... For Z(m), please refer to the previous introduction to Z(m) under the traditional CSI framework. This represents the processing latency information corresponding to the AI ​​function of the m-th CSI report, with the latency information in units of the uplink OFDM symbol. It is understandable that if the m-th CSI report is not actually determined based on the AI ​​function, i.e., it is not the first CSI report, then... It equals 0.

[0197] Similarly, the second delay For Z′(m), please refer to the previous introduction to Z′(m) under the traditional CSI framework.

[0198] In some embodiments, this application also provides a computer-readable storage medium including computer instructions that, when executed on an electronic device, cause a network device to perform the methods described above. The electronic device can be a terminal device or a network device.

[0199] In some embodiments, this application also provides a computer program product, including a computer program that, when executed by a processor, implements the method executed by the network device or the method executed by the terminal device as described above.

[0200] In some embodiments, this application also provides a communication device, including: a processor and a memory, wherein the memory is used to store program instructions, and the processor is used to execute the program instructions in the memory to implement the method performed by the network device as described above.

[0201] In some embodiments, this application also provides a communication device, including: a processor and a memory, wherein the memory is used to store program instructions, and the processor is used to execute the program instructions in the memory to implement the method performed by the terminal device as described above.

[0202] In some embodiments, this application also provides an electronic device that can function as a terminal device or a network device, the electronic device comprising: a memory and a processor. The memory and processor are coupled together. The memory stores computer program code, which includes computer instructions. The transceiver is used to receive and transmit data. When the processor executes the computer instructions, it causes the electronic device to perform the method described above.

[0203] In some embodiments, this application also provides a communication system, which may include terminal equipment and network equipment.

[0204] It is understood that any of the communication devices, terminal equipment, network equipment, electronic equipment, computer-readable storage media or computer program products provided above can be applied to the corresponding methods provided above. Therefore, the beneficial effects that can be achieved can be referred to the beneficial effects in the corresponding methods, and will not be repeated here.

[0205] These or other aspects of this application will become more readily apparent in the following description.

[0206] In addition, embodiments of this application also provide an apparatus, which may specifically be a chip, component or module. The apparatus may include a connected processor and a memory. The memory is used to store computer execution instructions. When the apparatus is running, the processor may execute the computer execution instructions stored in the memory to cause the chip to execute the communication methods executed by the network device or terminal device in the above method embodiments.

[0207] In this embodiment, the terminal device, computer storage medium, network device, computer program product or chip are all used to execute the corresponding methods provided above. Therefore, the beneficial effects that can be achieved can be referred to the beneficial effects of the corresponding methods provided above, and will not be repeated here.

[0208] Through the above description of the embodiments, those skilled in the art will understand that, for the sake of convenience and brevity, only the division of the above functional modules is used as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above.

[0209] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed. Furthermore, the mutual coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0210] The units described as separate components may or may not be physically separate. A component shown as a unit can be one or more physical units; that is, it can be located in one place or distributed in multiple different locations. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0211] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0212] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a readable storage medium. Based on this understanding, the technical solutions of the embodiments of this application, essentially or in other words, the parts that contribute to the prior art, or all or part of the technical solutions, can be embodied in the form of a software product. This software product is stored in a storage medium and includes several instructions to cause a device (which may be a microcontroller, chip, etc.) or processor to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0213] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A communication method, characterized in that, Applied to a terminal device, the method includes: The terminal device acquires first channel state information (CSI) capability information; wherein, the first CSI capability information includes the size of the neural network processing unit (NPU) and the size of the first CSI processing unit (CPU); the first CPU includes determining the CPU occupied by the first CSI report, or the first CPU includes determining the CPU occupied by the first CSI report and the second CSI report; the first CSI report represents a CSI report based on artificial intelligence (AI) functions, and the second CSI report represents a CSI report not based on AI functions; Send the first CSI capability information to the network device.

2. The method according to claim 1, characterized in that, The size of the NPU represents the number of NPUs shared by the AI ​​function groups supported by the terminal device; wherein, the AI ​​function group includes at least one AI function.

3. The method according to claim 1, characterized in that, The size of the NPU includes the number of NPUs corresponding to each AI function group supported by the terminal device; wherein, the AI ​​function group includes at least one AI function.

4. The method according to claim 2 or 3, characterized in that, The first CPU includes a CPU for determining a first CSI report and a second CSI report, and the size of the first CPU represents the number of CPUs shared by the AI ​​functions supported by the terminal device and the CPUs used to determine the second CSI report.

5. The method according to claim 3, characterized in that, The first CPU includes the CPU used to determine the first CSI report, and the size of the first CPU includes the number of CPUs corresponding to each AI function supported by the terminal device.

6. The method according to any one of claims 1 to 5, characterized in that, The first CSI capability information also includes AI processing latency information supported by the terminal device. The AI ​​processing latency information includes processing latency information corresponding to each AI function, processing latency information corresponding to each AI function group, or processing latency information corresponding to all AI functions. The AI ​​function group includes at least one AI function.

7. The method according to claim 6, characterized in that, The AI ​​processing latency information indicates the latency level or latency value.

8. The method according to any one of claims 1 to 7, characterized in that, The first CSI capability information also includes the AI ​​function group supported by the terminal device; wherein the AI ​​function group includes at least one AI function.

9. The method according to any one of claims 1 to 8, characterized in that, The method further includes: M CSI reports are determined from N CSI reports for updating; where N is a positive integer, 0≤M≤N, and M is an integer; the number of NPUs occupied by the M CSI reports is less than or equal to the number of unoccupied NPUs, and the number of unoccupied NPUs is determined based on the size of the NPUs and the number of occupied NPUs; the M CSI reports include a first CSI report and / or a second CSI report; N CSI reports are sent to the network device, and M of the N CSI reports are updated.

10. The method according to claim 9, characterized in that, The size of the NPU represents the number of NPUs shared by the AI ​​function groups supported by the terminal device, and the size of the first CPU represents the number of CPUs shared by the AI ​​functions supported by the terminal device and the CPUs used to determine the second CSI report. The N CSI reports include N AI The first CSI report, where 0≤N AI ≤N, and the N AI It is an integer; The number of NPUs occupied by the M1 first CSI reports in the M CSI reports is less than or equal to the number of unoccupied NPUs, and the number of CPUs occupied by the M CSI reports is less than or equal to the number of unoccupied CPUs. Where 0≤M1≤N AI M1 is an integer.

11. The method according to claim 10, characterized in that, The number of NPUs used by the M1 first CSI reports is greater than the number of NPUs used by the P1 first CSI reports, and the P1 first CSI reports include the N... AI At least one of the P1 first CSI reports, wherein the P1 first CSI reports are different from the M1 first CSI reports; Furthermore, the number of CPUs occupied by the M CSI reports is greater than the number of CPUs occupied by the P2 CSI reports, the P2 CSI reports include at least one of the N CSI reports, and the P2 CSI reports are different from the M CSI reports.

12. The method according to claim 10 or 11, characterized in that, The M CSI reports have a higher priority than the other CSI reports among the N CSI reports.

13. The method according to claim 9, characterized in that, The size of the NPU includes the number of NPUs corresponding to each AI function group supported by the terminal device, and the size of the first CPU represents all AI functions supported by the terminal device and the number of CPUs shared for determining the second CSI report. The N CSI reports include N AI The first CSI report, the N AI The first CSI report includes K AI functional groups, each corresponding to N. k The first CSI report; where 0≤N AI ≤N, N AI It is an integer, 1≤k≤K, where k is an integer; The M CSI reports include M2 ​​AI function groups, each corresponding to a Q. j The first CSI report states that the M2 AI function groups belong to the K AI function groups, where 1 ≤ j ≤ M2, and j is an integer. Q corresponding to the j-th AI function group j The number of NPUs used by the first CSI report is less than or equal to the number of unused NPUs corresponding to the j-th AI function group; The number of CPUs used by the M CSI reports is less than or equal to the number of CPUs not used.

14. The method according to claim 13, characterized in that, The Q corresponding to the j-th AI function group j The number of NPUs used by the first CSI report is greater than the number of NPUs used by P3 first CSI reports, where the P3 first CSI reports include the NPUs corresponding to the j-th AI function group. j At least one of the first CSI reports, the P3 first CSI reports and the Q j The first CSI report is different; Furthermore, the number of CPUs occupied by the M CSI reports is greater than the number of CPUs occupied by the P4 CSI reports, and the P4 CSI reports include at least one of the N CSI reports, and the P4 CSI reports are different from the M CSI reports.

15. The method according to claim 9, characterized in that, The size of the NPU includes the number of NPUs corresponding to each AI function group supported by the terminal device, and the size of the first CPU includes the number of CPUs corresponding to each AI function; the N CSI reports include N corresponding to K AI function groups. k The first CSI report; where 1 ≤ k ≤ K, and k is an integer; The step of determining M CSI reports from N CSI reports for updating includes: For each of the K AI function groups, from the N corresponding to the AI ​​function group k The first CSI report identified M k The first CSI report has been updated; The AI ​​function group corresponds to M k The number of NPUs used by each first CSI report is less than or equal to the number of unused NPUs corresponding to the AI ​​function group; the number of unused NPUs corresponding to the AI ​​function group is determined based on the number of NPUs corresponding to the AI ​​function group and the number of NPUs already used. And, the M corresponding to the AI ​​function group k The number of CPUs used by a first CSI report is less than or equal to the number of unused CPUs corresponding to the AI ​​function group; the number of unused CPUs corresponding to the AI ​​function group is determined based on the number of CPUs corresponding to the AI ​​function group and the number of CPUs already used.

16. The method according to claim 15, characterized in that, The AI ​​function group corresponds to M k The number of NPUs used by one first CSI report is greater than the number of NPUs used by P5 first CSI reports, where P5 first CSI reports include the NPUs corresponding to the AI ​​function group. k At least one of the first CSI reports, the P5 first CSI reports and the M k The first CSI report is different; And, the M corresponding to the AI ​​function group k The CPU usage of one first CSI report is greater than the CPU usage of P6 first CSI reports, where P6 first CSI reports include N corresponding to the AI ​​function group. k At least one of the first CSI reports, the P6 first CSI reports and the M k The first CSI report is different.

17. The method according to any one of claims 1 to 16, characterized in that, The CSI reference resource corresponding to the first CSI report is determined based on the first CSI reference resource and AI processing latency information; wherein, the first CSI reference resource is a traditional CSI reference resource.

18. The method according to any one of claims 1 to 16, characterized in that, When the time domain characteristic of the first CSI report is non-periodic, the first latency corresponding to the first CSI report is determined based on the traditional first latency and AI processing latency information, and the second latency corresponding to the first CSI report is determined based on the traditional second latency and AI processing latency information. Wherein, the first delay includes the minimum time for the terminal device to decode the physical downlink control channel (PDCCH) that triggers the first CSI report and to perform measurement calculations on the measurement reference signal; The second delay includes the minimum time required for the terminal device to perform measurement calculations on the measurement reference signal.

19. A communication method, characterized in that, Applied to network devices, the method includes: The receiving terminal device sends first channel state information (CSI) capability information; wherein, the first CSI capability information includes the size of the neural network processing unit (NPU) and the size of the first CSI processing unit (CPU); the first CPU includes determining the CPU occupied by the first CSI report, or the first CPU includes determining the CPU occupied by the first CSI report and the second CSI report; the first CSI report represents a CSI report based on artificial intelligence (AI) functions, and the second CSI report represents a CSI report not based on AI functions.

20. The method according to claim 19, characterized in that, The method further includes: The terminal device receives N CSI reports, of which M are updated CSI reports; where N is a positive integer, 0≤M≤N, and M is an integer; the number of NPUs occupied by the M CSI reports is less than or equal to the number of unoccupied NPUs, and the number of unoccupied NPUs is determined based on the size of the NPUs and the number of occupied NPUs; the M CSI reports include a first CSI report and / or a second CSI report.

21. A communication device, characterized in that, include: Module for performing the communication method as described in any one of claims 1-18; And / or, a module for performing the communication method as described in any one of claims 19-20.

22. A communication system, characterized in that, include: Terminal equipment and network equipment; The terminal device acquires first channel state information (CSI) capability information; wherein, the first CSI capability information includes the size of the neural network processing unit (NPU) and the size of the first CSI processing unit (CPU); the first CPU includes determining the CPU occupied by the first CSI report, or the first CPU includes determining the CPU occupied by the first CSI report and the second CSI report; the first CSI report represents a CSI report based on artificial intelligence functions, and the second CSI report represents a CSI report not based on AI functions; The terminal device sends the first CSI capability information to the network device; The network device receives the first CSI capability information.

23. A communication device, characterized in that, include: At least one processor and an interface circuit, the interface circuit being configured to receive signals from other communication devices besides the communication device and transmit them to the processor or to send signals from the processor to other communication devices besides the communication device, the processor being configured to implement the method as described in any one of claims 1-18 via logic circuits or executable code instructions; and / or, the processor being configured to implement the method as described in any one of claims 19-20 via logic circuits or executable code instructions.

24. A computer-readable storage medium, characterized in that, It includes a computer program or instructions that, when run on a computer, cause the computer to perform the method as described in any one of claims 1-18; and / or cause the computer to perform the method as described in any one of claims 19-20.

25. A chip, characterized in that, include: An interface circuit and a logic circuit, wherein the interface circuit is used to receive signals from other chips besides the chip and transmit them to the logic circuit, or to send signals from the logic circuit to other chips besides the chip, and the logic circuit is used to implement the method as described in any one of claims 1-18; and / or, the logic circuit is used to implement the method as described in any one of claims 19-20.

26. A computer program product, characterized in that, The computer program product includes: a computer program or instructions that, when executed on a computer, cause the computer to perform the method as described in any one of claims 1-18; and / or cause the computer to perform the method as described in any one of claims 19-20.

Images