CSI Reporting Method, Apparatus, Device, and Storage Medium

Abstract

The present disclosure provides a CSI reporting method, apparatus, device, and storage medium. Here, the method includes transmitting a channel state information (CSI) report to a network device, the CSI report including N sets of reporting amounts, each set of reporting amounts including an index of at least one reference signal and a layer 1 reference signal received power (L1-RSRP) or a layer 1 signal-to-interference-plus-noise ratio (L1-SINR) corresponding to the at least one reference signal, where N represents the number of sets of reporting amounts set by the network device, and N is a positive integer.

Description

Technical Field

[0001] The present disclosure relates to the field of wireless communication technologies, and in particular, to a CSI reporting method, its apparatus, device, and storage medium.

[0002] (Cross-reference to related applications) This application claims priority to an application with application number 202210553989.3, filed with the China National Intellectual Property Administration on May 20, 2022, and all the contents of the said application are incorporated herein by reference.

Background Art

[0003] Currently, the reporting of Channel State Information (CSI) includes two reporting methods. The first is the normal reporting method, that is, when groupBasedBeamReporting, a parameter group configured by network devices, is set to Disable, the terminal reports two or four reference signals. The second is the group-based reporting method, that is, when groupBasedBeamReporting configured by network devices is set to Enable, the terminal reports two reference signals at a time. However, both CSI reporting methods are inefficient.

Summary of the Invention

Problems to be Solved by the Invention

[0004] In view of this, embodiments of the present disclosure aim to provide a CSI reporting method, its apparatus, device, and storage medium.

Means for Solving the Problems

[0005] The technical solutions of the embodiments of the present disclosure are realized as follows.

[0006] At least one embodiment of the present disclosure provides a CSI reporting method executed by a terminal, and the method includes: transmitting a CSI report to a network device, where the CSI report includes N sets of reporting quantities, and each set of reporting quantities includes an index of at least one reference signal and a layer one-reference signal receive power (L1-RSRP) or a layer one-signal to interference plus noise ratio (L1-SINR) corresponding to the at least one reference signal; here, N represents the number of sets of reporting quantities set by the network device, and N is a positive integer.

[0007] Furthermore, according to at least one embodiment of the present disclosure, a slot applied to at least one reference signal in each set of reporting quantities and the L1-RSRP or L1-SINR corresponding to the at least one reference signal is determined based on a first period set by the network device.

[0008] Furthermore, according to at least one embodiment of the present disclosure, the method further includes: quantizing the maximum L1-RSRP value or the maximum L1-SINR value in the N sets of reporting quantities with K bits, using the maximum L1-RSRP value or the maximum L1-SINR value as a reference, calculating each of the other L1-RSRP values or L1-SINR values in the N sets of reporting quantities excluding the maximum L1-RSRP value or the maximum L1-SINR value as a difference to obtain a differential L1-RSRP value or a differential L1-SINR value, where the differential L1-RSRP value or the differential L1-SINR value is quantized with P bits, and here, both K and P are positive integers, and K is greater than P.

[0009] Furthermore, according to at least one embodiment of the present disclosure, the CSI report further includes indication information, and the indication information indicates position information of at least one L1-RSRP or L1-SINR included in at least one of N reporting amount sets of the maximum L1-RSRP value or the maximum L1-SINR value.

[0010] Furthermore, according to at least one embodiment of the present disclosure, the method further includes performing non-uniform quantization on the L1-RSRP value or the L1-SINR value corresponding to each reference signal in the N reporting amount sets by a first artificial intelligence (AI) module.

[0011] Furthermore, according to at least one embodiment of the present disclosure, the total number of bits for performing non-uniform quantization on the L1-RSRP value or the L1-SINR value corresponding to each reference signal in the N reporting amount sets is smaller than a first threshold value.

[0012] Furthermore, according to at least one embodiment of the present disclosure, the method further includes obtaining auxiliary information transmitted from the network device, and the auxiliary information includes channel state information and / or scheduling information.

[0013] Furthermore, according to at least one embodiment of the present disclosure, the method further includes transmitting related information of the first AI module to the network device.

[0014] Furthermore, according to at least one embodiment of the present disclosure, the method Among the M sets of reporting amounts, when there are a plurality of sets of reporting amounts that meet the predetermined conditions, one set of reporting amounts is retained from the plurality of sets of reporting amounts, and among the plurality of sets of reporting amounts, other sets of reporting amounts except for the one retained set of reporting amounts are excluded from the M sets of reporting amounts to obtain N sets of reporting amounts, where M is a positive integer, M is equal to N, or M is greater than N.

[0015] Furthermore, according to at least one embodiment of the present disclosure, the CSI report further includes the index of at least one reference signal in the other excluded sets of reporting amounts, or the CSI report further includes the L1-RSRP or L1-SINR corresponding to at least one reference signal in the other excluded sets of reporting amounts.

[0016] Furthermore, according to at least one embodiment of the present disclosure, meeting the predetermined conditions includes that the indexes of the reference signals for every two sets of reporting amounts in the plurality of sets of reporting amounts are all the same or partially the same, or for every two sets of reporting amounts in the plurality of sets of reporting amounts, calculating the ratios of the L1-RSRP or L1-SINR corresponding to each reference signal in one set of reporting amounts of the two sets of reporting amounts to the L1-RSRP or L1-SINR corresponding to each reference signal in the other set of reporting amounts to obtain a plurality of ratios, and all the plurality of ratios are smaller than a second threshold, or for every two sets of reporting amounts in the plurality of sets of reporting amounts, calculating the differences between the L1-RSRP or L1-SINR corresponding to each reference signal in one set of reporting amounts of the two sets of reporting amounts and the L1-RSRP or L1-SINR corresponding to each reference signal in the other set of reporting amounts to obtain a plurality of differences, and all the plurality of differences are smaller than a second threshold.

[0017] Embodiments of the present disclosure provide a CSI reporting method executed by a network device, and the method includes Receiving a CSI report transmitted from a terminal, where the CSI report includes N sets of reporting quantities, and each set of reporting quantities includes an index of at least one reference signal and L1-RSRP or L1-SINR corresponding to the at least one reference signal, Here, N represents the number of sets of reporting quantities set by the network device, and N is a positive integer.

[0018] Furthermore, according to at least one embodiment of the present disclosure, the slot applied to at least one reference signal in each set of reporting quantities and L1-RSRP or L1-SINR corresponding to the at least one reference signal is determined based on a first period set by the network device.

[0019] Furthermore, according to at least one embodiment of the present disclosure, the method further includes setting the number N of sets of reporting quantities and the first period for the terminal.

[0020] Furthermore, according to at least one embodiment of the present disclosure, the CSI report further includes indication information, and the indication information indicates the position information of the maximum L1-RSRP value or the maximum L1-SINR value in at least one L1-RSRP or L1-SINR included in the N sets of reporting quantities.

[0021] Furthermore, according to at least one embodiment of the present disclosure, the method further includes restoring the L1-RSRP or L1-SINR of each reference signal of the N sets of non-uniformly quantized reporting quantities by a second AI module.

[0022] Furthermore, according to at least one embodiment of the present disclosure, the total number of bits for performing non-uniform quantization on the L1-RSRP value or the L1-SINR value corresponding to each reference signal in the N sets of reporting quantities is smaller than a first threshold.

[0023] Furthermore, according to at least one embodiment of the present disclosure, the method further comprising transmitting auxiliary information to the terminal, wherein the auxiliary information includes channel state information and / or scheduling information.

[0024] Furthermore, according to at least one embodiment of the present disclosure, the method further comprises receiving related information of a first AI module transmitted from the terminal.

[0025] Furthermore, according to at least one embodiment of the present disclosure, when there are multiple reporting amount sets satisfying a predetermined condition among the M reporting amount sets, retaining one reporting amount set from the multiple reporting amount sets, excluding other reporting amount sets except the retained one reporting amount set from the M reporting amount sets to obtain N reporting amount sets, where M is a positive integer, M is equal to N, or M is greater than N.

[0026] Furthermore, according to at least one embodiment of the present disclosure, the CSI report further includes indexes of at least one reference signal in other excluded reporting amount sets, or the CSI report further includes L1-RSRP or L1-SINR corresponding to at least one reference signal in other excluded reporting amount sets.

[0027] Furthermore, according to at least one embodiment of the present disclosure, satisfying the predetermined condition means that indexes of reference signals for every two reporting amount sets in the multiple reporting amount sets are all the same or partially the same, or for every two reporting amount sets in the multiple reporting amount sets, calculating ratios of L1-RSRP or L1-SINR corresponding to each reference signal in one of the two reporting amount sets to L1-RSRP or L1-SINR corresponding to each reference signal in the other reporting amount set respectively to obtain a plurality of ratios, and all the plurality of ratios are smaller than a second threshold value, or For each pair of two reporting quantity sets within the plurality of reporting quantity sets, calculate the difference between the L1-RSRP or L1-SINR corresponding to each reference signal in one of the two reporting quantity sets and the L1-RSRP or L1-SINR corresponding to each reference signal in the other reporting quantity set, respectively, to obtain a plurality of differences, and all of the plurality of differences are smaller than a second threshold value.

[0028] Embodiments of the present disclosure provide a CSI reporting apparatus, and the apparatus includes a transmission unit configured to transmit a CSI report to a network device, where the CSI report includes N reporting quantity sets, and each reporting quantity set includes an index of at least one reference signal and an L1-RSRP or L1-SINR corresponding to the at least one reference signal, where N represents the number of reporting quantity sets set by the network device, and N is a positive integer.

[0029] Embodiments of the present disclosure provide a CSI reporting apparatus, and the apparatus includes a receiving unit configured to receive a CSI report transmitted from a terminal, where the CSI report includes N reporting quantity sets, and each reporting quantity set includes an index of at least one reference signal and an L1-RSRP or L1-SINR corresponding to the at least one reference signal, where N represents the number of reporting quantity sets set by the network device, and N is a positive integer.

[0030] At least one embodiment of the present disclosure provides a terminal including a first processor and a first memory storing a computer program executable by the processor, where the first processor executes all processes in any one of the above methods on the terminal side by executing the computer program.

[0031] At least one embodiment of the present disclosure provides a network device including a second processor and a second memory storing a computer program executable by the processor. Here, the second processor executes all processes in any one of the above methods on the network device side by executing the computer program.

[0032] At least one embodiment of the present disclosure provides a storage medium storing a computer program for causing a processor to execute all processes in any one of the above methods.

Advantages of the Invention

[0033] According to the CSI reporting method, apparatus, device, and storage medium provided by the embodiments of the present disclosure, a network device transmits a CSI report, the CSI report includes N sets of reporting quantities, each set of reporting quantities includes at least one index of a reference signal and L1-RSRP or L1-SINR corresponding to the at least one reference signal, where N represents the number of sets of reporting quantities set by the network device, and N is a positive integer. According to the technical solution of the embodiments of the present disclosure, a terminal can improve CSI reporting efficiency by reducing the number of reports by transmitting all reference signals waiting for reporting to the network device at one time in the form of sets of reporting quantities.

Brief Description of the Drawings

[0034]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

DETAILED DESCRIPTION OF THE INVENTION

[0035] Before introducing the technical solutions according to the embodiments of the present disclosure, the related art will be first described.

[0036] In the related art, in downlink beam reporting (i.e., CSI reporting (CSI Reporting)) based on layer one-reference signal receive power (L1-RSRP), the CSI reporting is set by a CSI-report configuration (CSI-ReportConfig).

[0037] Table 1 shows an overview of reporting beams by adopting two reporting methods in the related art. As shown in Table 1, specifically, it may include the following two reporting methods. The first one is a normal reporting method in which groupBasedBeamReporting is set to Disable and the terminal reports two or four reference signals. The second one is a group-based reporting method in which groupBasedBeamReporting is set to Enable and two synchronization signal block (SSB) resource indicators (SSBRI) or two channel state information reference signal (CSI-RS) resource indicators (CRI) are reported at a time. The beams corresponding to these two SSBRI / CRI are two beams that can be received simultaneously. The efficiency of any CSI reporting method is low.

[0038]

Table 1

[0039] FIG. 1 is a diagram showing that a terminal in the related art reports two beams at a time. As shown in FIG. 1, in the case of a multi-panel or wide beam, the base station can transmit services to the terminal by using these two beams simultaneously, improving the multi-stream transmission ability of the high-frequency analog beam.

[0040] Currently, in the 5G design framework of related technologies, AI can gain significant benefits in many fields such as demodulation reference signal (DMRS) detection, CSI-RS overhead reduction, CSI feedback, beam management, and position measurement, showing excellent application potential. The research content of wireless AI can include three use cases of CSI feedback, beam management, and position measurement, as well as methods for AI model deployment, inference, update, and simulation evaluation. Here, a typical use case of beam management includes beam prediction in the time domain and spatial domain to reduce overhead and delay and improve beam selection accuracy. Table 2 shows an overview of the SID of the use cases of air interface AI.

[0041]

Table 2

[0042] In the related art, the following technical problems exist in the beam reporting of CSI. First, in the CSI reporting framework, in each CSI report, it is only possible to report a maximum of four beams and the corresponding L1-RSRP / L1-SINR. Second, the reported beam index and the corresponding L1-RSRP / L1-SINR are based on the CSI-RS / SSB transmitted by the base station in the past, that is, it is not possible to report the L1-RSRP or L1-SINR within a certain period in the future. Third, in a scenario with high-frequency and high-speed movement, the time-varying nature of the channel is large and the beam changes quickly, so the beam failure probability is high. To adapt to such a scenario, the CSI reporting framework requires more frequent beam measurements and beam reports. As a result, the base station has additional overhead for reference signals and CSI reports, and the terminal also has additional overhead for measurements and reports, and the capacity and interference of the reference signals in the network are also more severely tested. Fourth, in the time-series-based L1-RSRP or L1-SINR reporting scheme, it is necessary to feedback nrofTimeDomainBeamReporting (abbreviated as N) × nrofReportedRS (abbreviated as M) SSBRI / CRIs in one CSI report. For each SSBRI / CRI, it is necessary to report the corresponding L1-RSRP / L1-SINR. For L1-RSRP, the L1-RSRP within the range of [-140, -44] dBm is indicated by 7 bits, and the step size is 1 dB. For L1-SINR, the L1-SINR within the range of [-23, 40] dBm is indicated by 7 bits, and the step size is 0.5 dB. In one feedback, 7×N×M bits are required to carry L1-RSRP / L1-SINR, which may impose a significant uplink channel overhead on the system. Therefore, how to carry the L1-RSRP / L1-SINR required for CSI reporting with fewer bits has become an urgent issue to be solved.

[0043] In view of this, in an embodiment of the present disclosure, the terminal transmits a CSI report to the network device, the CSI report includes N sets of reporting amounts, each set of reporting amounts includes an index of at least one reference signal and L1-RSRP or L1-SINR corresponding to the at least one reference signal, where N represents the number of sets of reporting amounts set by the network device, and N is a positive integer.

[0044] FIG. 2 is an exemplary flowchart of the implementation of the CSI reporting method according to an embodiment of the present disclosure, which is executed by a terminal. As shown in FIG. 2, the method includes step 201.

[0045] In step 201, a CSI report is transmitted to the network device, the CSI report includes N sets of reporting amounts, each set of reporting amounts includes an index of at least one reference signal and L1-RSRP or L1-SINR corresponding to the at least one reference signal, where N represents the number of sets of reporting amounts set by the network device, and N is a positive integer.

[0046] It can be understood that the reference signal includes a beam. The index refers to SSBRI / CRI.

[0047] It can be understood that when the terminal reports a plurality of reference signals to the network device, the terminal can transmit all the reference signals waiting to be reported to the network device at one time in the form of a set of reporting amounts.

[0048] For example, assuming that N is 2, the first set of reporting amounts includes an index of one reference signal and the corresponding L1-RSRP or L1-SINR, and the second set of reporting amounts includes indexes of two reference signals and the corresponding L1-RSRP or L1-SINR. The specific forms of the two sets of reporting amounts can be as follows. {CRI_0, L1-RSRP0}, {CRI_1, L1-RSRP1, CRI_2, L1-RSRP2}. or {CRI_0, L1-SINR0}, {CRI_1, L1-SINR1, CRI_2, L1-SINR2}

[0049] In actual application, the terminal can indicate the reference signal that the network device will use in a certain future period. That is, the terminal can predict the reference signal used by the network device within a plurality of fixed periods after the time when the terminal transmits the CSI report.

[0050] Based on this, in one embodiment, the slot applied to at least one reference signal in each reporting amount set and the L1-RSRP or L1-SINR corresponding to the at least one reference signal is determined based on a first period set by the network device.

[0051] Here, the first period may refer to the CSI reporting time domain period.

[0052] Here, if the CSI reporting time domain period is indicated by PeriodofTimeDomainBeamReporting, the slot applied to at least one reference signal in the i-th reporting amount set can be determined using i×PeriodofTimeDomainBeamReporting.

[0053] For example, there are two sets of reported quantities. The first set of reported quantities includes the index of one reference signal and the corresponding L1-RSRP or L1-SINR. The second set of reported quantities includes the indices of two reference signals and the corresponding L1-RSRP or L1-SINR. Assuming that the CSI reporting time domain period PeriodofTimeDomainBeamReporting = 5 ms, the slot applied to the reference signal and the corresponding L1-RSRP or L1-SINR in the first set of reported quantities is indicated by 1×5 ms = 5 ms. That is, at the first time (1×5 ms = 5 ms), the reference signal and the corresponding L1-RSRP or L1-SINR in the first set of reported quantities are used. The slots applied to the two reference signals and the corresponding L1-RSRP or L1-SINR in the second set of reported quantities are both indicated by 2×5 ms = 10 ms. That is, at the second time (2×5 ms = 10 ms), the two reference signals and the L1-RSRP or L1-SINR in the second set of reported quantities are used. The specific forms of the two sets of reported quantities can be shown as follows. {CRI_0,0, L1-RSRP0}, {CRI_1,1, L1-RSRP1, CRI_1,2, L1-RSRP2}, or {CRI_0,0, L1-SINR0}, {CRI_1,1, L1-SINR1, CRI_1,2, L1-SINR2}.

[0054] Here, CRI_i,j indicates the reference signal j at time i×PeriodofTimeDomainBeamReporting, where i ∈ {0, …, nrofTimeDomainBeamReporting - 1} and j ∈ {0, …, nrofReportedRS - 1}.

[0055] Understandably, the network device can send a CSI reporting configuration to the terminal to set the number N of reporting amount sets and the first period for the terminal. The network device can introduce the number N of reporting amount sets and the first period into the CSI-ReportConfig IE of RRC. The number of reporting amount sets can be indicated by the nrofTimeDomainBeamReporting parameter, and the first period can be indicated by the PeriodofTimeDomainBeamReporting parameter. The terminal can send N reporting amount sets to the network device by adopting one of the following methods.

[0056] The first method is normal reporting.

[0057] Specifically, when the parameter groupBasedBeamReporting introduced by the network device into the CSI-ReportConfig IE of RRC is set to Disable, the terminal can report nrofTimeDomainBeamReporting (N) reporting amount sets in one report. Each reporting amount set contains nrofReportedRS (M) SSBRI / CRIs, and the i-th reporting amount set is applied to the i × PeriodofTimeDomainBeamReporting slots after the CSI reporting time.

[0058] For example, when the parameter groupBasedBeamReporting introduced by the network device into the CSI-ReportConfig IE of RRC is set to Disable, the network device sets nrofTimeDomainBeamReporting = 4 and nrofReportedRS = 2 in the CSI-ReportConfig of RRC. The four reporting amount sets that the terminal can report can specifically be as follows. {CRI_0,0, L1-RSRP0, CRI_0,1, L1-RSRP1}, {CRI_1,0, L1-RSRP0, CRI_1,1, L1-RSRP1}, {CRI_2,0, L1-RSRP0, CRI_2,1, L1-RSRP1}, {CRI_3,0, L1-RSRP0, CRI_3,1, L1-RSRP1, CRI_3,2, L1-RSRP2}。

[0059] Here, CRI_i,j represents the reference signal j at time i×PeriodofTimeDomainBeamReporting, where i∈{0,…,nrofTimeDomainBeamReporting - 1} and j∈{0,…,nrofReportedRS - 1}.

[0060] The second method is group - based reporting.

[0061] Specifically, when the parameter groupBasedBeamReporting introduced by the network device in the RRC's CSI - ReportConfig IE is set to Enable, the terminal can report nrofTimeDomainBeamReporting (N) reporting quantity sets in one report. Each reporting quantity set contains two SSBRI / CRI, and the beams corresponding to these two SSBRI / CRI are two beams that can be received simultaneously. For example, in the case of multi - panel or wide beam, the network device, such as a base station, can use these two beams to transmit services to the terminal simultaneously, thereby improving the multi - stream transmission ability of high - frequency analog beams.

[0062] For example, when the parameter groupBasedBeamReporting introduced by the network device in the RRC's CSI - ReportConfig IE is set to Enable, the network device sets nrofTimeDomainBeamReporting = 2 in the RRC's CSI - ReportConfig. The two reporting quantity sets that the terminal can report can be specifically as follows. {CRI_0,0, L1-RSRP0, CRI_0,1, L1-RSRP1} {CRI_1,0, L1-RSRP0, CRI_1,1, L1-RSRP1}

[0063] Here, CRI_i,j represents the reference signal j at time i×PeriodofTimeDomainBeamReporting, where i∈{0,…,nrofTimeDomainBeamReporting - 1} and j∈{0,…,nrofReportedRS - 1}.

[0064] In the following, how to save the bit overhead of the report will be described for each case.

[0065] In the case of the first case, differential quantization is used to save the bit overhead of the report.

[0066] In actual applications, considering that the numerical ranges of L1-RSRP or L1-SINR of each reference signal are different, if the same number of bits is used for quantization of L1-RSRP or L1-SINR of each reference signal in N reporting quantity sets, the bit overhead will increase. Therefore, the maximum L1-RSRP or L1-SINR can be selected, and the difference between each of the other L1-RSRPs and the maximum L1-RSRP can be obtained to get a difference value, or the difference between each of the other L1-SINRs and the maximum L1-SINR can be obtained to get a difference value. In this way, relatively large bits can be used to perform quantization on the maximum L1-RSRP or L1-SINR, and relatively small bits can be used to perform quantization on the difference value.

[0067] Based on this, in one embodiment, the method is Quantize the maximum L1-RSRP value or the maximum L1-SINR value in N sets of reported quantities with K bits, and calculate each L1-RSRP value or L1-SINR value other than the maximum L1-RSRP value or the maximum L1-SINR value in the N sets of reported quantities as a difference with reference to the maximum L1-RSRP value or the maximum L1-SINR value, further including obtaining a differential L1-RSRP value or a differential L1-SINR value, where the differential L1-RSRP value or the differential L1-SINR value is quantized with P bits. Here, both K and P are positive integers, and K is greater than P.

[0068] For example, assuming N is 2, the first set of reported quantities includes the index of one reference signal and the corresponding L1-RSRP or L1-SINR, and the second set of reported quantities includes the indices of two reference signals and the corresponding L1-RSRP or L1-SINR. The two sets of reported quantities can be specifically as follows. {CRI_0, L1-SINR0}, {CRI_1, L1-SINR1, CRI_2, L1-SINR2}.

[0069] Here, assuming the maximum L1-SINR is L1-SINR0, K = 7, and P = 4, calculate the difference values between L1-SINR1 and L1-SINR0, and between L1-SINR2 and L1-SINR0 to obtain two differential L1-SINR values. Assuming L1-SINR0 is quantized with 7 bits and is 0000111, and the two differential L1-SINR values are both quantized with 4 bits and are 0110 and 0101, then two quantized sets of reported quantities are obtained.

[0070] In actual application, in order for the network device to be able to restore the corresponding L1-RSRP or L1-SINR, the terminal can indicate to the network device the position of at least one L1-RSRP or L1-SINR in the N reporting amount sets of the maximum L1-RSRP or maximum L1-SINR. Thereby, the network device can restore the maximum L1-RSRP value or maximum L1-SINR value based on K bits and can restore the remaining L1-RSRP or L1-SINR using P bits.

[0071] Based on this, in one embodiment, the CSI report further includes indication information, and the indication information indicates position information of at least one L1-RSRP or L1-SINR in the N reporting amount sets of the maximum L1-RSRP value or maximum L1-SINR value.

[0072] For example, assume that the network device sets nrofTimeDomainBeamReporting = N and nrofReportedRS = M in the CSI-ReportConfig of RRC. The terminal reports N reporting amount sets, and each reporting amount set includes M SSBRI / CRIs and the corresponding L1-RSRP or L1-SINR. The maximum L1-RSRP or L1-SINR is quantized with K bits, and the K bits may refer to the L1-RSRP or L1-SINR within a specific dBm range. The other (N×M)-1 L1-RSRPs or L1-SINRs are quantized with P bits, and the P bits indicate the difference value between the other L1-RSRP and the maximum L1-RSRP or the difference value between the other L1-SINR and the maximum L1-SINR. Here, P < K.

[0073] Furthermore, the terminal uses an indication information bitmap with a length of log(N×M) to indicate the position of at least one L1-RSRP or L1-SINR included in N reporting amount sets of the maximum L1-RSRP or the maximum L1-SINR. In this way, the network device can restore with K bits based on the indication information bitmap to obtain the maximum L1-RSRP value or the maximum L1-SINR value, and restore with P bits to obtain other L1-RSRP or L1-SINR.

[0074] For example, in one report, N×M (N = 2, M = 4) SSBRI / CRI and the corresponding L1-RSRP / L1-SINR are required. Assuming that N×M = 2×4 = 8 L1-RSRP are reported, the maximum L1-RSRP is quantized with 7 bits, and the 7 bits indicate the L1-RSRP within the range of [-140, -44]) dBm, the step size is 1 dB, and a bitmap with a length of 3 bits can indicate the position of the maximum L1-RSRP in 8 L1-RSRP. The other 7 L1-RSRP are quantized with 4 bits, and the 4 bits can indicate the difference value from the maximum L1-RSRP, and the step size is 2 dB.

[0075] Assuming that N×M = 2×4 = 8 L1-SINR are reported, the maximum L1-SINR is quantized with 7 bits, indicating the L1-SINR within the range of [-23, 40] dBm, the step size is 1 dB, and a bitmap with a length of 3 bits indicates the position of the maximum L1-SINR in 8 L1-SINR. The other 7 L1-SINR are quantized with 4 bits, and the 4 bits can indicate the difference value from the maximum L1-SINR, and the step size is 2 dB.

[0076] Table 3 shows that differential quantization is used to save the bit overhead of reporting. As shown in Table 3, it is assumed that the network device sets that the terminal reports N CSI reports, reports the index of M reference signals and the corresponding L1-RSRP or L1-SINR each time, and the network device transmits B CSI-RS or SSB each time.

[0077]

Table 3

[0078] In the case of the second case, non-uniform quantization is used to save the bit overhead of the report.

[0079] In actual applications, since AI can obtain great gains in many fields such as DMRS detection, CSI-RS overhead reduction, CSI feedback, beam management, and position measurement, and shows excellent application potential, a first AI module can be provided in the terminal, and the first AI module performs non-uniform quantization on the L1-RSRP value or L1-SINR value of each reference signal of N sets of reported amounts. A second AI module can be provided in the network device, and the second AI module restores the L1-RSRP or L1-SINR of each reference signal of the N sets of reported amounts that have been non-uniformly quantized.

[0080] Based on this, in one embodiment, the method further includes: performing non-uniform quantization on the L1-RSRP value or L1-SINR value corresponding to each reference signal in the N sets of reported amounts by the first AI module.

[0081] It can be understood that the non-uniform quantization may refer to the fact that the number of bits for quantizing the L1-RSRP or L1-SINR of each reference signal in the N sets of reported amounts is different.

[0082] It can be understood that based on the numerical range of the L1-RSRP or L1-SINR of each reference signal, the number of bits used for quantizing the L1-RSRP or L1-SINR can be determined. For example, for the L1-RSRP or L1-SINR within the first numerical range, the first bit is adopted for quantization, and for the L1-RSRP or L1-SINR within the second numerical range, the second bit is adopted for quantization.

[0083] For example, assuming that N is 2, the first set of reported quantities includes the index of one reference signal and the corresponding L1-SINR, and the second set of reported quantities includes the indices of two reference signals and the corresponding L1-SINR. The two sets of reported quantities may specifically be as follows. {CRI_0, L1-SINR0}, {CRI_1, L1-SINR1, CRI_2, L1-SINR2}.

[0084] Here, the terminal determines, by means of a trained quantization algorithm, i.e., the first AI module, the number of bits (assumed to be 7 bits) used for quantizing L1-SINR0 based on the numerical range of L1-SINR0, determines the number of bits (assumed to be 4 bits) used for quantizing L1-SINR1 based on the numerical range of L1-SINR1, and determines the number of bits (assumed to be 2 bits) used for quantizing L1-SINR2 based on the numerical range of L1-SINR2.

[0085] Furthermore, the terminal reports three unquantized L1-SINRs to the network device. The network device restores the first L1-SINR0 quantized with 7 bits, the second L1-SINR1 quantized with 4 bits, and the third L1-SINR2 quantized with 2 bits by means of a trained dequantization algorithm, i.e., the second AI module.

[0086] In one embodiment, the total number of bits for performing non-uniform quantization on the L1-RSRP value or L1-SINR value corresponding to each reference signal within N sets of reported quantities is smaller than the first threshold.

[0087] For example, the total number of bits for performing non-uniform quantization on the L1-RSRP or L1-SINR values of N×M reference signals by the terminal using the first AI module is Abits, where A is smaller than the first threshold. Here, A may be set by the RRC transmitted from the network device, or A may be a fixed determined value.

[0088] In actual application, the terminal can perform non-uniform quantization on the L1-RSRP value or L1-SINR value of each reference signal in N reporting amount sets by the first AI module, and send the reporting amount set obtained by the non-uniform quantization to the network device. Thus, the network device can restore the L1-RSRP or L1-SINR of each reference signal in the reporting amount set obtained by the non-uniform quantization by the second AI module. To ensure that the terminal and the network device use the corresponding AI modules, the network device can send auxiliary information to the terminal and expect that the terminal can select an AI module based on the auxiliary information. Thereby, the terminal can use the AI module matching the auxiliary information to realize non-uniform quantization of L1-RSRP or L1-SINR, and the network device can use the AI module matching the auxiliary information to restore the non-uniformly quantized L1-RSRP or L1-SINR.

[0089] Based on this, in one embodiment, the method further includes obtaining the auxiliary information sent from the network device, where the auxiliary information includes channel state information and / or scheduling information.

[0090] Here, after obtaining the auxiliary information, the terminal can select an AI module matching the auxiliary information from a predetermined database, and use the selected AI module matching the auxiliary information as the first AI module.

[0091] FIG. 3 is a diagram showing that a terminal and a network device perform non-uniform quantization and restoration on the L1-RSRP value or the L1-SINR value using their respective AI modules. As shown in FIG. 3, the terminal performs non-uniform quantization on the L1-RSRP or L1-SINR values of N×M reference signals using a trained quantization algorithm, i.e., an AI encoding module. The network device restores the non-uniform quantized L1-RSRP or L1-SINR value using a trained quantization cancellation algorithm, i.e., an AI decoding module.

[0092] For example, assuming that the terminal reports two L1-RSRP or L1-SINR values, the terminal quantizes the first L1-RSRP or L1-SINR value to 1 bit and the second L1-RSRP or L1-SINR value to 2 bits by a trained quantization algorithm, i.e., the first AI module, and the terminal reports the two unquantized L1-RSRP or L1-SINR values to the network device. The network device restores the first L1-RSRP or L1-SINR value quantized to 1 bit and the second L1-RSRP or L1-SINR value quantized to 2 bits by a trained quantization cancellation algorithm, i.e., the second AI module.

[0093] It can be understood that the network device can further instruct the terminal to use one or more of the AI modules based on changes in the channel state and channel overhead. Here, the non-uniform quantization process among multiple AI models is not affected, and only the inference efficiency of the terminal is affected.

[0094] It can be understood that when the network device instructs the AI module, it can also further instruct related information such as the accuracy, complexity, and calculation time of the AI module. Thereby, the terminal can select the corresponding first AI module from the predetermined database based on the related information.

[0095] In actual application, the terminal can further customize and select a first AI module, and send the related information of the selected first AI module to the network device. In this way, the network device uses a second AI module corresponding to the first AI module to restore the non-uniformly quantized L1-RSRP or L1-SINR.

[0096] Based on this, in one embodiment, the method further includes: sending the related information of the selected first AI module to the network device.

[0097] Here, the terminal can randomly select one AI module from a predetermined database as the first AI module, and send the related information of the selected first AI module to the network device. The related information may include related information such as accuracy, complexity, and calculation time. The predetermined database may store a plurality of AI modules.

[0098] Understandably, the terminal can report the related information of one or more AI modules selected by itself to the network device via MAC CE or UCI based on its own state.

[0099] Understandably, after receiving the related information of the first module sent from the terminal, the network device can use the related information to determine a second AI module that matches the first AI module.

[0100] Understandably, the first AI module can be deployed on the terminal side, and the second AI module can be deployed on the network device side. Specifically, the AI module may refer to the training and inference modules of AI / machine learning (ML), and training and inference parameters such as the weights of the convolutional layers can also be interacted between the terminal and the network device.

[0101] In order to ensure that the terminal and the network device use the corresponding AI modules, the terminal can send the relevant information of the first AI module to the network device, expecting that the network device can select an AI module based on the relevant information. Thereby, the terminal can use the first AI module to realize non-uniform quantization of L1-RSRP or L1-SINR, and the network device can use the second AI module matching the relevant information to restore the non-uniformly quantized L1-RSRP or L1-SINR.

[0102] Table 4 shows that the terminal saves the bit overhead of the report by non-uniform quantization. As shown in Table 4, it is assumed that the network device sets that the terminal reports N CSI reports, reports the index of M reference signals and the corresponding L1-RSRP or L1-SINR each time, and the network device transmits B CSI-RS or SSB each time.

[0103]

Table 4

[0104] In the case of the third case, the bit overhead of the report is saved by reducing the number of reported L1-RSRP or L1-SINR.

[0105] In actual applications, considering that the terminal can predict the reference signals to be used by the network device in a certain future period, when the indexes of the reference signals used by the network device predicted at multiple fixed periods in the future are the same, for multiple fixed periods, only the index of one reference signal and the corresponding L1-RSRP or L1-SINR may be reported. Or, when the L1-RSRP or L1-SINR of the reference signals used by the network device predicted at multiple fixed periods in the future are the same, or when the sizes of the L1-RSRP or L1-SINR of the reference signals used by the network device predicted at multiple fixed periods in the future are similar, for multiple fixed periods, only the index of one reference signal and the corresponding L1-RSRP or L1-SINR may be reported. That is, in the form of a set of reporting amounts, conditional screening is performed on multiple reference signals waiting to be reported to reduce the number of L1-RSRP or L1-SINR to be reported.

[0106] Based on this, in one embodiment, the method includes When there are multiple reporting amount sets that meet the predetermined conditions among the M reporting amount sets further including retaining one reporting amount set from the multiple reporting amount sets, and excluding the other reporting amount sets except the retained one reporting amount set from the M reporting amount sets to obtain N reporting amount sets, where M is a positive integer, M is equal to N, or M is greater than N.

[0107] It can be understood that meeting the predetermined conditions means that the indexes of the reference signals for every two reporting amount sets in the multiple reporting amount sets are all the same or partially the same, or For each pair of two reporting quantity sets within the plurality of reporting quantity sets, calculate the ratio of the L1-RSRP or L1-SINR corresponding to each reference signal in one of the two reporting quantity sets to the L1-RSRP or L1-SINR corresponding to each reference signal in the other reporting quantity set, respectively, to obtain a plurality of ratios, and all of the plurality of ratios are smaller than a second threshold value. Or For each pair of two reporting quantity sets within the plurality of reporting quantity sets, calculate the difference between the L1-RSRP or L1-SINR corresponding to each reference signal in one of the two reporting quantity sets and the L1-RSRP or L1-SINR corresponding to each reference signal in the other reporting quantity set, respectively, to obtain a plurality of differences, and all of the plurality of differences are smaller than a second threshold value, which is included.

[0108] In one embodiment, the CSI report further includes the index of at least one reference signal in another reporting quantity set that is excluded, or the CSI report further includes the L1-RSRP or L1-SINR corresponding to at least one reference signal in another reporting quantity set that is excluded.

[0109] In addition, when reporting the index of at least one reference signal in one excluded reporting quantity set, the index of at least one reference signal in the reporting quantity set can be set as a specific value, indicating that the index indicated at a previous time or a later time at this moment is adopted. Or, when reporting the L1-RSRP or L1-SINR corresponding to at least one reference signal in one excluded reporting quantity set, the L1-RSRP or L1-SINR corresponding to at least one reference signal in the reporting quantity set is set as a specific value, indicating that the L1-RSRP or L1-SINR indicated at a previous time or a later time at this moment is adopted.

[0110] In Example 1, the network device sets nrofTimeDomainBeamReporting = 4 and nrofReportedRS = 2 in the CSI-ReportConfig of RRC. The four sets of reported quantities calculated by the terminal are as follows. {CRI_0,0, CRI_0,1, L1-SINR0, L1-SINR1}, {CRI_0,0, CRI_0,1, L1-SINR0, L1-SINR1}, {CRI_2,0, CRI_2,1, L1-SINR0, L1-SINR1}, {CRI_3,0, CRI_3,1, L1-SINR0, L1-SINR1}.

[0111] Here, since the indexes of the reference signals in the first set of reported quantities and the second set of reported quantities are all the same, one of the two sets of reported quantities is retained, and the other set of reported quantities is excluded from the four sets of reported quantities. The three sets of reported quantities to be reported are specifically as follows. {CRI_0,0, CRI_0,1, L1-SINR0, L1-SINR1}, {CRI_2,0, CRI_2,1, L1-SINR0, L1-SINR1}, {CRI_3,0, CRI_3,1, L1-SINR0, L1-SINR1}, Or {CRI_0,0, CRI_0,1}, {CRI_0,0, CRI_0,1, L1-SINR0, L1-SINR1}, {CRI_2,0, CRI_2,1, L1-SINR0, L1-SINR1}, {CRI_3,0, CRI_3,1, L1-SINR0, L1-SINR1}, Or {CRI_0,0, CRI_0,1, L1-SINR0, L1-SINR1}, {CRI_0,0, CRI_0,1}, {CRI_2,0, CRI_2,1, L1-SINR0, L1-SINR1}, {CRI_3,0, CRI_3,1, L1-SINR0, L1-SINR1}, or {L1-SINR0, L1-SINR1}, {CRI_0,0, CRI_0,1, L1-SINR0, L1-SINR1}, {CRI_2,0, CRI_2,1, L1-SINR0, L1-SINR1}, {CRI_3,0, CRI_3,1, L1-SINR0, L1-SINR1}, or {CRI_0,0, CRI_0,1, L1-SINR0, L1-SINR1}, {L1-SINR0, L1-SINR1}, {CRI_2,0, CRI_2,1, L1-SINR0, L1-SINR1}, {CRI_3,0, CRI_3,1, L1-SINR0, L1-SINR1}.

[0112] In Example 2, the network device sets nrofTimeDomainBeamReporting = 4 and nrofReportedRS = 1 in the CSI-ReportConfig of RRC. The four sets of reported quantities calculated by the terminal are as follows respectively. {CRI_0,0, L1-SINR0}, {CRI_0,0, L1-SINR0}, {CRI_2,0, L1-SINR0}, {CRI_3,0, L1-SINR0}.

[0113] Here, since the indexes of the reference signals in the first set of reported quantities and the second set of reported quantities are all the same, one of the two sets of reported quantities is retained, and the other set of reported quantities is excluded from the four sets of reported quantities. The three sets of reported quantities to be reported are specifically as follows. {CRI_0,0, L1-SINR0}, {CRI_2,0, L1-SINR0}, {CRI_3,0, L1-SINR0}, Or {CRI_0,0}, {CRI_0,0, L1-SINR0}, {CRI_2,0, L1-SINR0}, {CRI_3,0, L1-SINR0}, Or {L1-SINR0}, {CRI_0,0, L1-SINR0}, {CRI_2,0, L1-SINR0}, {CRI_3,0, L1-SINR0}, Or {CRI_0,0, L1-SINR0}, {CRI_0,0}, {CRI_2,0, L1-SINR0}, {CRI_3,0, L1-SINR0}, Or {CRI_0,0, L1-SINR0}, {L1-SINR0}, {CRI_2,0, L1-SINR0}, {CRI_3,0, L1-SINR0}.

[0114] In Example 3, the network device sets nrofTimeDomainBeamReporting = 4 and nrofReportedRS = 2 in the CSI-ReportConfig of RRC, and the four beam groups calculated by the terminal are as follows respectively. {CRI_0,0, CRI_0,1, L1-SINR0, L1-SINR1}, {CRI_1,0, CRI_0,1, L1-SINR0, L1-SINR1}, {CRI_2,0, CRI_2,1, L1-SINR0, L1-SINR1}, {CRI_3,0, CRI_3,1, L1-SINR0, L1-SINR1}.

[0115] Here, since some of the indexes of the reference signals in the first reported quantity set and the second reported quantity set are the same, one of the two reported quantity sets is retained, and the other reported quantity set is excluded from the four reported quantity sets. The three reported quantity sets to be reported are specifically as follows. {CRI_0,0, CRI_0,1, L1 - SINR0, L1 - SINR1}, {CRI_2,0, CRI_2,1, L1 - SINR0, L1 - SINR1}, {CRI_3,0, CRI_3,1, L1 - SINR0, L1 - SINR1}, Or {CRI_0,0, CRI_0,1, L1 - SINR0, L1 - SINR1}, {CRI_1,0, CRI_0,1}, {CRI_2,0, CRI_2,1, L1 - SINR0, L1 - SINR1}, {CRI_3,0, CRI_3,1, L1 - SINR0, L1 - SINR1}, Or {CRI_0,0, CRI_0,1, L1 - SINR0, L1 - SINR1}, {L1 - SINR0, L1 - SINR1}, {CRI_2,0, CRI_2,1, L1 - SINR0, L1 - SINR1}, {CRI_3,0, CRI_3,1, L1 - SINR0, L1 - SINR1}, Or {CRI_1,0, CRI_0,1, L1 - SINR0, L1 - SINR1}, {CRI_2,0, CRI_2,1, L1 - SINR0, L1 - SINR1}, {CRI_3,0, CRI_3,1, L1 - SINR0, L1 - SINR1}, Or {CRI_0,0, CRI_0,1}, {CRI_1,0, CRI_0,1, L1 - SINR0, L1 - SINR1}, {CRI_2,0, CRI_2,1, L1 - SINR0, L1 - SINR1}, {CRI_3,0, CRI_3,1, L1-SINR0, L1-SINR1}, or {L1-SINR0, L1-SINR1}, {CRI_1,0, CRI_0,1, L1-SINR0, L1-SINR1}, {CRI_2,0, CRI_2,1, L1-SINR0, L1-SINR1}, {CRI_3,0, CRI_3,1, L1-SINR0, L1-SINR1}.

[0116] In Example 4, the network device sets nrofTimeDomainBeamReporting = 4 and nrofReportedRS = 2 in the CSI-ReportConfig of RRC. The four beam groups calculated by the terminal are as follows. {CRI_0,0, CRI_0,1, L1-SINR0, L1-SINR1}, {CRI_1,0, CRI_1,1, L1-SINR0, L1-SINR1}, {CRI_2,0, CRI_2,1, L1-SINR0, L1-SINR1}, {CRI_3,0, CRI_3,1, L1-SINR0, L1-SINR1}.

[0117] Here, taking the first reporting amount set and the second reporting amount set as examples, calculate the corresponding ratios of the L1-SINR of each reference signal in the first reporting amount set and the second reporting amount set to obtain two ratios, or calculate the corresponding differences of the L1-SINR of each reference signal in the first reporting amount set and the second reporting amount set to obtain two difference values. If both of the two ratios are smaller than the second threshold value, or both of the two difference values are smaller than the third threshold value, one of the two reporting amount sets can be retained and the other reporting amount set can be excluded from the four reporting amount sets. The three reported reporting amount sets are specifically as follows. {CRI_0,0, CRI_0,1, L1-SINR0, L1-SINR1}, {CRI_2,0, CRI_2,1, L1 - SINR0, L1 - SINR1}, {CRI_3,0, CRI_3,1, L1 - SINR0, L1 - SINR1}, or {CRI_0,0, CRI_0,1, L1 - SINR0, L1 - SINR1}, {CRI_1,0, CRI_1,1}, {CRI_2,0, CRI_2,1, L1 - SINR0, L1 - SINR1}, {CRI_3,0, CRI_3,1, L1 - SINR0, L1 - SINR1}, or {CRI_0,0, CRI_0,1, L1 - SINR0, L1 - SINR1}, {L1 - SINR0, L1 - SINR1}, {CRI_2,0, CRI_2,1, L1 - SINR0, L1 - SINR1}, {CRI_3,0, CRI_3,1, L1 - SINR0, L1 - SINR1}, or {CRI_1,0, CRI_1,1, L1 - SINR0, L1 - SINR1}, {CRI_2,0, CRI_2,1, L1 - SINR0, L1 - SINR1}, {CRI_3,0, CRI_3,1, L1 - SINR0, L1 - SINR1}, or {CRI_0,0, CRI_0,1}, {CRI_1,0, CRI_1,1, L1 - SINR0, L1 - SINR1}, {CRI_2,0, CRI_2,1, L1 - SINR0, L1 - SINR1}, {CRI_3,0, CRI_3,1, L1 - SINR0, L1 - SINR1}, or {L1 - SINR0, L1 - SINR1}, {CRI_1,0, CRI_1,1, L1 - SINR0, L1 - SINR1}, {CRI_2,0, CRI_2,1, L1 - SINR0, L1 - SINR1}, {CRI_3,0, CRI_3,1, L1 - SINR0, L1 - SINR1}。

[0118] In addition, when reporting the index of at least one reference signal or L1 - RSRP or L1 - SINR within one set of reported quantities that has been excluded, the index of at least one reference signal within the set of reported quantities can be set as a specific value, indicating that the index or L1 - RSRP or L1 - SINR instructed at the previous time or the subsequent time at this moment is adopted.

[0119] Table 5 shows that the reporting bit overhead can be saved by reducing the number of L1 - RSRP or L1 - SINR reported by the terminal. As shown in Table 5, it is assumed that the network device sets that the terminal reports N times of CSI reports, reports the index of M reference signals and the corresponding L1 - RSRP or L1 - SINR each time, and the network device transmits B CSI - RS or SSB each time.

[0120]

Table 5

[0121] In the embodiments of the present disclosure, the following advantages are achieved.

[0122] (1) The terminal can improve the CSI reporting efficiency by reducing the number of reports by transmitting all the reference signals waiting to be reported to the network device at one time in the form of a set of reported quantities.

[0123] (2) The terminal can predict the index of the reference signals and the corresponding L1 - RSRP or L1 - SINR that the network device will use in a certain future period, and report the index of the reference signals and the corresponding L1 - RSRP or L1 - SINR that the network device will use in the predicted certain future period in one CSI report.

[0124] (3) By means such as differential quantization, non-uniform quantization, and reducing the number of reported L1-RSRP or L1-SINR, the bit overhead of reported L1-RSRP or L1-SINR can be reduced, the CSI reporting overhead can be reduced, and the beam accuracy can be improved.

[0125] FIG. 4 is an exemplary flowchart of the realization of the CSI reporting method according to an embodiment of the present disclosure, which is executed by a network device. As shown in FIG. 4, the method includes step 401.

[0126] In step 401, a CSI report transmitted from a terminal is received. The CSI report includes N sets of reported quantities. Each set of reported quantities includes at least one reference signal index and the L1-RSRP or L1-SINR corresponding to the at least one reference signal. Here, N represents the number of sets of reported quantities set by the network device, and N is a positive integer.

[0127] Understandably, the reference signal includes a beam. The index refers to SSBRI / CRI.

[0128] Understandably, when the terminal reports a plurality of reference signals to the network device, the terminal can transmit all the reference signals waiting to be reported to the network device at one time in the form of a set of reported quantities.

[0129] In actual application, the terminal can indicate the reference signal that the network device will use in a certain future period. That is, the terminal can predict the reference signal that the network device will use within a plurality of fixed periods after the time when the terminal transmits the CSI report.

[0130] Based on this, in one embodiment, the slot applied to at least one reference signal in each of the reported quantity sets and the L1-RSRP or L1-SINR corresponding to the at least one reference signal is determined based on a first period set by the network device.

[0131] Here, the first period may refer to the CSI reporting time domain period.

[0132] In actual application, the network device can send a CSI reporting configuration to the terminal to set the number N of the reported quantity sets and the first period for the terminal.

[0133] Based on this, in one embodiment, the method further includes: setting the number N of the reported quantity sets and the first period for the terminal.

[0134] In actual application, considering that the numerical ranges of the L1-RSRP or L1-SINR of each reference signal are different, if the same bits are used for quantization of the L1-RSRP or L1-SINR of each reference signal in N reported quantity sets, the bit overhead will increase. Therefore, the terminal can select the maximum L1-RSRP or L1-SINR, obtain the differences between the remaining L1-RSRP or L1-SINR and the maximum L1-RSRP or L1-SINR to obtain difference values. In this way, relatively large bits can be used to perform quantization on the maximum L1-RSRP or L1-SINR, and relatively small bits can be used to perform quantization on the difference values. Further, in order for the network device to be able to restore the corresponding L1-RSRP or L1-SINR, the terminal can indicate to the network device the positions of the reference signals corresponding to the maximum L1-RSRP or maximum L1-SINR in the N reported quantity sets. Thereby, the network device can restore the maximum L1-RSRP value or maximum L1-SINR value based on K bits and restore the remaining L1-RSRP or L1-SINR using P bits.

[0135] Based on this, in one embodiment, the CSI report further includes indication information, and the indication information indicates the position information of at least one L1-RSRP or L1-SINR included in N reporting quantity sets of the maximum L1-RSRP value or the maximum L1-SINR value.

[0136] For example, assume that the network device sets nrofTimeDomainBeamReporting = N and nrofReportedRS = M in the CSI-ReportConfig of RRC. The terminal reports N reporting quantity sets, and each reporting quantity set includes M SSBRI / CRIs and corresponding L1-RSRP or L1-SINR. The maximum L1-RSRP or L1-SINR is quantized in K bits, and the K bits may refer to the L1-RSRP or L1-SINR within a specific dBm range. The other (N×M)-1 L1-RSRPs or L1-SINRs are quantized in P bits, and the P bits indicate the difference value between the other L1-RSRP and the maximum L1-RSRP, or the difference value between the other L1-SINR and the maximum L1-SINR. Here, P < K.

[0137] Furthermore, the terminal uses an indication information bitmap with a length of log(N×M) to indicate the position of at least one L1-RSRP or L1-SINR included in N reporting quantity sets of the maximum L1-RSRP or the maximum L1-SINR. In this way, the network device can restore the maximum L1-RSRP value or the maximum L1-SINR value in K bits and restore the other L1-RSRP or L1-SINR in P bits based on the indication information bitmap.

[0138] In actual applications, since AI can achieve great gains in many fields such as DMRS detection, CSI-RS overhead reduction, CSI feedback, beam management, and position measurement, showing excellent application potential, a first AI module can be provided in the terminal, and the first AI module performs non-uniform quantization on the L1-RSRP value or L1-SINR value of each reference signal in N sets of reporting quantities. A second AI module can be provided in the network device, and the second AI module restores the L1-RSRP or L1-SINR of each reference signal in the N sets of reporting quantities that have been non-uniformly quantized.

[0139] Based on this, in one embodiment, the method further includes restoring, by the second AI module, the L1-RSRP or L1-SINR of each reference signal in the N sets of reporting quantities that have been non-uniformly quantized.

[0140] For example, assuming N is 2, the first set of reporting quantities includes the index of one reference signal and the corresponding L1-SINR, and the second set of reporting quantities includes the indexes of two reference signals and the corresponding L1-SINR. The two sets of reporting quantities may specifically be as follows. {CRI_0, L1-SINR0}, {CRI_1, L1-SINR1, CRI_2, L1-SINR2}.

[0141] Here, the terminal determines, based on the numerical range of L1-SINR0, the number of bits (assumed to be 7 bits) used for quantization of L1-SINR0, determines, based on the numerical range of L1-SINR1, the number of bits (assumed to be 4 bits) used for quantization of L1-SINR1, and determines, based on the numerical range of L1-SINR2, the number of bits (assumed to be 2 bits) used for quantization of L1-SINR2 by using a trained quantization algorithm, that is, the first AI module.

[0142] Furthermore, the terminal reports three non-quantized L1-SINRs to the network device. The network device restores the first L1-SINR0 quantized in 7 bits, the second L1-SINR1 quantized in 4 bits, and the third L1-SINR2 quantized in 2 bits by a trained dequantization algorithm, i.e., the second AI module.

[0143] In one embodiment, the total number of bits for performing non-uniform quantization on the L1-RSRP value or L1-SINR value corresponding to each reference signal in the N reporting amount sets is smaller than the first threshold.

[0144] In actual applications, the terminal can perform non-uniform quantization on the L1-RSRP value or L1-SINR value of each reference signal in the N reporting amount sets by the first AI module, and transmit the reporting amount set obtained by non-uniform quantization to the network device. Thereby, the network device can restore the L1-RSRP or L1-SINR of each reference signal in the reporting amount set obtained by non-uniform quantization by the second AI module. To ensure that the terminal and the network device use the corresponding AI modules, it can be expected that the network device transmits auxiliary information to the terminal so that the terminal can select an AI module based on the auxiliary information. Thereby, the terminal can use the AI module matching the auxiliary information to realize non-uniform quantization of L1-RSRP or L1-SINR, and the network device can use the AI module matching the auxiliary information to restore the non-uniformly quantized L1-RSRP or L1-SINR.

[0145] Based on this, in one embodiment, the method further includes transmitting auxiliary information to the terminal, where the auxiliary information includes channel state information and / or scheduling information.

[0146] In actual applications, the terminal can further customize and select a first AI module and send the related information of the selected first AI module to the network device. In this way, the network device uses a second AI module corresponding to the first AI module to restore the non-uniformly quantized L1-RSRP or L1-SINR.

[0147] Based on this, in one embodiment, the method further includes: receiving the related information of the first AI module sent from the terminal.

[0148] Here, the terminal can select one AI module from a predetermined database as the first AI module and send the related information of the selected first AI module to the network device. The related information may include related information such as accuracy, complexity, and calculation time. The predetermined database may store a plurality of AI modules.

[0149] Note that in order to ensure that the terminal and the network device use the corresponding AI modules, the terminal can send the related information of the first AI module to the network device, expecting the network device to be able to select an AI module based on the related information. Thereby, the terminal can use the first AI module to realize the non-uniform quantization of L1-RSRP or L1-SINR, and the network device can use the second AI module matching the related information to restore the non-uniformly quantized L1-RSRP or L1-SINR.

[0150] In actual applications, considering that the terminal can predict the reference signals that the network device will use in a certain future period, when the indexes of the reference signals used by the network device predicted at multiple fixed periods in a certain future period are the same, for multiple fixed periods, only the index of one reference signal and the corresponding L1-RSRP or L1-SINR may be reported. Or, when the L1-RSRP or L1-SINR of the reference signals used by the network device predicted at multiple fixed periods in a certain future period are the same, or when the sizes of the L1-RSRP or L1-SINR of the reference signals used by the network device predicted at multiple fixed periods in a certain future period are similar, for multiple fixed periods, only the index of one reference signal and the corresponding L1-RSRP or L1-SINR may be reported. That is, in the form of a set of reporting amounts, by performing conditional screening on multiple reference signals waiting to be reported, the number of reported L1-RSRP or L1-SINR is reduced.

[0151] Based on this, in one embodiment, among the M sets of reporting amounts, if there are multiple sets of reporting amounts that meet the predetermined conditions, one set of reporting amounts is retained from the multiple sets of reporting amounts, and among the multiple sets of reporting amounts, other sets of reporting amounts except the retained one set of reporting amounts are excluded from the M sets of reporting amounts to obtain N sets of reporting amounts.

[0152] Here, M is a positive integer, M is equal to N, or M is greater than N.

[0153] It can be understood that meeting the predetermined conditions means in the multiple sets of reporting amounts, the indexes of the reference signals for every two sets of reporting amounts are all the same or partially the same, Or For each pair of two reporting amount sets among the plurality of reporting amount sets, calculate the ratio of the L1-RSRP or L1-SINR corresponding to each reference signal in one of the two reporting amount sets to the L1-RSRP or L1-SINR corresponding to each reference signal in the other reporting amount set, respectively, to obtain a plurality of ratios, and all of the plurality of ratios are smaller than a second threshold value. Or For each pair of two reporting amount sets among the plurality of reporting amount sets, calculate the difference between the L1-RSRP or L1-SINR corresponding to each reference signal in one of the two reporting amount sets and the L1-RSRP or L1-SINR corresponding to each reference signal in the other reporting amount set, respectively, to obtain a plurality of differences, and all of the plurality of differences are smaller than a second threshold value, which is included.

[0154] In one embodiment, the CSI report further includes the index of at least one reference signal in another reporting amount set that has been excluded. Or The CSI report further includes the L1-RSRP or L1-SINR corresponding to at least one reference signal in another reporting amount set that has been excluded.

[0155] Note that when reporting the index of at least one reference signal in one excluded reporting amount set, the index of at least one reference signal in the reporting amount set can be set as a specific value, indicating that the index indicated at a previous time or a later time is adopted at this time. Or, when reporting the L1-RSRP or L1-SINR corresponding to at least one reference signal in one excluded reporting amount set, the L1-RSRP or L1-SINR corresponding to at least one reference signal in the reporting amount set is set as a specific value, indicating that the L1-RSRP or L1-SINR indicated at a previous time or a later time is adopted at this time.

[0156] In the embodiments of the present disclosure, the following advantages are achieved.

[0157] (1) The terminal can improve the CSI reporting efficiency by reducing the number of reports by transmitting all the reference signals waiting to be reported to the network device at one time in the reporting amount set method.

[0158] (2) The terminal can predict the indexes of the reference signals to be used by the network device in a future certain period and the corresponding L1-RSRP or L1-SINR, and report, in one CSI report, the indexes of the reference signals to be used by the network device in the predicted future certain period and the corresponding L1-RSRP or L1-SINR.

[0159] (3) By means such as differential quantization, non-uniform quantization, and reducing the L1-RSRP or L1-SINR to be reported, the bit overhead of the L1-RSRP or L1-SINR to be reported can be reduced, the CSI reporting overhead can be reduced, and the beam accuracy can be improved.

[0160] To implement the CSI reporting method of the embodiments of the present disclosure, the embodiments of the present disclosure further provide a CSI reporting apparatus. FIG. 5 is an exemplary structural diagram of the configuration of the CSI reporting apparatus of the embodiments of the present disclosure. As shown in FIG. 5, the apparatus includes a transmission unit 51 configured to transmit a CSI report to the network device. The CSI report includes N reporting amount sets, and each reporting amount set includes at least one index of a reference signal and the L1-RSRP or L1-SINR corresponding to the at least one reference signal. Here, N represents the number of reporting amount sets set by the network device, and N is a positive integer.

[0161] In one embodiment, the slots applied to at least one reference signal in each reporting amount set and the L1-RSRP or L1-SINR corresponding to the at least one reference signal are determined based on a first period set by the network device.

[0162] In one embodiment, the apparatus further includes Quantize the maximum L1-RSRP value or the maximum L1-SINR value in N sets of reported amounts with K bits, and use the maximum L1-RSRP value or the maximum L1-SINR value as a reference to calculate each L1-RSRP value or L1-SINR value other than the maximum L1-RSRP value or the maximum L1-SINR value in the N sets of reported amounts as a difference to obtain a differential L1-RSRP value or a differential L1-SINR value, where the differential L1-RSRP value or the differential L1-SINR value is quantized with P bits, where K and P are both positive integers, and K is greater than P.

[0163] In one embodiment, the CSI report further includes indication information, and the indication information indicates the position information of the maximum L1-RSRP value or the maximum L1-SINR value in at least one L1-RSRP or L1-SINR included in the N sets of reported amounts.

[0164] In one embodiment, the apparatus further is configured by a first AI module to perform non-uniform quantization on the L1-RSRP value or the L1-SINR value corresponding to each reference signal in N sets of reported amounts.

[0165] Furthermore, according to at least one embodiment of the present disclosure, the total number of bits for performing non-uniform quantization on the L1-RSRP value or the L1-SINR value corresponding to each reference signal in N sets of reported amounts is smaller than a first threshold.

[0166] In one embodiment, the apparatus further is configured to obtain auxiliary information transmitted from the network device, and the auxiliary information includes channel state information and / or scheduling information.

[0167] Furthermore, according to at least one embodiment of the present disclosure, the method further includes transmitting related information of the first AI module to the network device.

[0168] In one embodiment, the apparatus further Among the M sets of reported quantities, when there are multiple sets of reported quantities that meet the predetermined conditions, one set of reported quantities is retained from the multiple sets of reported quantities, and other sets of reported quantities except the retained one set of reported quantities among the multiple sets of reported quantities are excluded from the M sets of reported quantities to obtain N sets of reported quantities, and is configured to where M is a positive integer, M is equal to N, or M is greater than N.

[0169] In one embodiment, the CSI report further includes an index of at least one reference signal in the other excluded sets of reported quantities, or the CSI report further includes L1-RSRP or L1-SINR corresponding to at least one reference signal in the other excluded sets of reported quantities.

[0170] In one embodiment, meeting the predetermined conditions means that the indexes of the reference signals for every two sets of reported quantities in the multiple sets of reported quantities are all the same or partially the same, or for every two sets of reported quantities in the multiple sets of reported quantities, the ratios of L1-RSRP or L1-SINR corresponding to each reference signal in one of the two sets of reported quantities to L1-RSRP or L1-SINR corresponding to each reference signal in the other set of reported quantities are calculated respectively to obtain a plurality of ratios, and all of the plurality of ratios are smaller than a second threshold, or for every two sets of reported quantities in the multiple sets of reported quantities, the differences between L1-RSRP or L1-SINR corresponding to each reference signal in one of the two sets of reported quantities and L1-RSRP or L1-SINR corresponding to each reference signal in the other set of reported quantities are calculated respectively to obtain a plurality of differences, and all of the plurality of differences are smaller than a second threshold, and includes.

[0171] In actual applications, the transmission unit 51 can be implemented by a communication interface in the CSI reporting device.

[0172] When the CSI reporting device according to the above embodiment performs CSI reporting, only the division of each of the above program modules is taken as an example for explanation. However, in actual applications, the above processing may be assigned to different program modules and performed as needed. That is, the internal structure of the device can be divided into different program modules to perform all or part of the above processing. Furthermore, the CSI reporting device according to the above embodiment belongs to the same concept as the embodiment of the CSI reporting method. The specific implementation process can refer to the embodiment of the method and will not be repeatedly described here.

[0173] To implement the CSI reporting method of the embodiments of the present disclosure, the embodiments of the present disclosure further provide a CSI reporting device. FIG. 6 is an exemplary structural diagram of the configuration of the CSI reporting device according to the embodiments of the present disclosure. As shown in FIG. 6, the device includes a receiving unit 61 configured to receive a CSI report transmitted from a terminal, where the CSI report includes N sets of reporting quantities, and each set of reporting quantities includes at least one index of a reference signal and L1-RSRP or L1-SINR corresponding to the at least one reference signal. Here, N represents the number of sets of reporting quantities set by the network device, and N is a positive integer.

[0174] In one embodiment, the slots applied to at least one reference signal and L1-RSRP or L1-SINR corresponding to the at least one reference signal in each set of reporting quantities are determined based on a first period set by the network device.

[0175] In one embodiment, the device further includes a configuration to set the number N of sets of reporting quantities and the first period for the terminal.

[0176] In one embodiment, the CSI report further includes indication information, and the indication information indicates position information of at least one L1-RSRP or L1-SINR among N sets of reporting amounts of the maximum L1-RSRP value or the maximum L1-SINR value.

[0177] In one embodiment, the apparatus further is configured by a second AI module to restore the L1-RSRP or L1-SINR of each reference signal of N sets of non-uniformly quantized reporting amounts.

[0178] In one embodiment, the total number of bits for performing non-uniform quantization on the L1-RSRP value or the L1-SINR value corresponding to each reference signal in N sets of reporting amounts is smaller than a first threshold value.

[0179] In one embodiment, the apparatus further is configured to transmit auxiliary information to the terminal, and the auxiliary information includes channel state information and / or scheduling information.

[0180] In one embodiment, the apparatus further is configured to receive related information of the first AI module transmitted from the terminal.

[0181] In one embodiment, when there are a plurality of reporting amount sets satisfying a predetermined condition among M reporting amount sets, one reporting amount set is retained from the plurality of reporting amount sets, and other reporting amount sets except the retained one reporting amount set among the plurality of reporting amount sets are excluded from the M reporting amount sets to obtain N reporting amount sets, where M is a positive integer, M is equal to N, or M is greater than N.

[0182] In one embodiment, the CSI report further includes indexes of at least one reference signal in the other excluded reporting amount sets, or The CSI report further includes L1-RSRP or L1-SINR corresponding to at least one reference signal in other sets of reported quantities that are excluded.

[0183] In one embodiment, satisfying the predetermined condition means that in the plurality of sets of reported quantities, the indexes of the reference signals for every two sets of reported quantities are all the same or partially the same, or for every two sets of reported quantities in the plurality of sets of reported quantities, calculate the ratios of L1-RSRP or L1-SINR corresponding to each reference signal in one of the two sets of reported quantities to L1-RSRP or L1-SINR corresponding to each reference signal in the other set of reported quantities, respectively, to obtain a plurality of ratios, and all of the plurality of ratios are smaller than a second threshold value, or for every two sets of reported quantities in the plurality of sets of reported quantities, calculate the differences between L1-RSRP or L1-SINR corresponding to each reference signal in one of the two sets of reported quantities and L1-RSRP or L1-SINR corresponding to each reference signal in the other set of reported quantities, respectively, to obtain a plurality of differences, and all of the plurality of differences are smaller than a second threshold value, which is included.

[0184] In actual applications, the receiving unit 61 can be realized by a communication interface in the CSI reporting device.

[0185] Note that when the CSI reporting device according to the above embodiment performs CSI reporting, only the division of each of the above program modules is taken as an example for description. However, in actual applications, the above processing may be assigned to different program modules and executed as needed, that is, the internal structure of the device is divided into different program modules to execute all or part of the above processing. Furthermore, the CSI reporting device according to the above embodiment belongs to the same concept as the embodiment of the CSI reporting method, and its specific realization process can refer to the embodiment of the method, and will not be described repeatedly here.

[0186] Embodiments of the present disclosure further provide a terminal. As shown in FIG. 7, the terminal includes: a first communication interface 71 capable of interacting with other devices for information; a first processor 72 connected to the first communication interface 71 and configured to execute a computer program to implement the method according to one or more technical solutions on the terminal side. The computer program is stored in a first memory 73.

[0187] It should be noted that for the specific processing procedures of the first processor 72 and the first communication interface 71, reference may be made to the method embodiments and will not be repeated here.

[0188] Of course, in actual applications, each component in the terminal 70 is coupled by a bus system 74. It is understandable that the bus system 74 is used to realize connection and communication between these components. In addition to the data bus, the bus system 74 further includes a power bus, a control bus, and a status signal bus. However, for clarity of description, in FIG. 7, various buses are denoted as the bus system 74.

[0189] The first memory 73 in the embodiments of the present disclosure is configured to store various types of data to support the operation of the terminal 70. Examples of these data include any computer program operated on the terminal 70.

[0190] The method disclosed in the embodiments of the present disclosure can be applied to or implemented by the first processor 72. The first processor 72 can be an integrated circuit chip having signal processing capabilities. In the implementation process, each step of the above method can be performed by the integrated logic circuit of the hardware in the first processor 72 or instructions in the form of software. The above-mentioned first processor 72 can be a general-purpose processor, a digital signal processor (DSP), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The first processor 72 can implement or execute each method, step, and logic block diagram disclosed in the embodiments of the present disclosure. The general-purpose processor can be a microprocessor or any conventional processor, etc. The steps of the method disclosed in the embodiments of the present disclosure may be directly executed by a hardware decoding processor, or may be executed by a combination of hardware and software modules in the decoding processor. The software module can be arranged in a storage medium, and the storage medium is arranged in the first memory 73. The first processor 72 reads the information in the first memory 73 and combines it with its hardware to perform the steps of the above method.

[0191] Embodiments of the present disclosure further provide a network device. As shown in FIG. 8, the network device includes a second communication interface 81 capable of interacting with other devices for information, and a second processor 82 connected to the second communication interface 81 and configured to execute a method by one or more technical solutions on the network device side by executing a computer program. The computer program is stored in the second memory 83.

[0192] Note that for the specific processing processes of the second processor 82 and the second communication interface 81, reference can be made to the method embodiments and will not be repeatedly described here.

[0193] Of course, in actual applications, each component within the network device 80 is coupled by a bus system 84. Understandably, the bus system 84 is used to realize the connection communication between these components. In addition to the data bus, the bus system 84 further includes a power bus, a control bus, and a status signal bus. However, for clarity of explanation, in FIG. 8, various buses are denoted as the bus system 84.

[0194] The second memory 83 in the embodiments of the present disclosure is configured to store various types of data to support the operation of the network device 80. Examples of these data include any computer program operated on the network device 80.

[0195] The method disclosed in the above embodiments of the present disclosure can be applied to or implemented by the second processor 82. The second processor 82 can be an integrated circuit chip having signal processing capabilities. In the implementation process, each step of the above method can be performed by the integrated logic circuit of the hardware in the second processor 82 or instructions in the form of software. The above-mentioned second processor 82 can be a general-purpose processor, a digital signal processor (DSP), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The second processor 82 can implement or execute each method, step and logic block diagram disclosed in the embodiments of the present disclosure. The general-purpose processor can be a microprocessor or any conventional processor, etc. The steps of the method disclosed in the embodiments of the present disclosure may be directly executed by a hardware decoding processor, or may be executed by a combination of hardware and software modules in the decoding processor. The software module can be arranged in a storage medium, and the storage medium is arranged in the second memory 83. The second processor 82 reads the information in the second memory 83 and combines it with its hardware to perform the steps of the above method.

[0196] In an exemplary embodiment, the terminal 70 and the network device 80 can be implemented by one or more application specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), complex programmable logic devices (CPLDs), field-programmable gate arrays (FPGAs), general-purpose processors, controllers, microcontroller units (MCUs), microprocessors, or other electronic elements and can be used to execute the above method.

[0197] Understandably, the memories (the first memory 73 and the second memory 83) of the embodiments of the present disclosure may be volatile memories or non-volatile memories, or may include both volatile memories and non-volatile memories. Here, the non-volatile memory can be a read-only memory (ROM: Read Only Memory), a programmable read-only memory (PROM: Programmable Read-Only Memory), an erasable programmable read-only memory (EPROM: Erasable Programmable Read-Only Memory), an electrically erasable programmable read-only memory (EEPROM: Electrically Erasable Programmable Read-Only Memory), a ferromagnetic random access memory (FRAM (registered trademark): Ferromagnetic Random Access Memory), a flash memory (Flash Memory), a magnetic surface memory, an optical disc, or a compact disc read-only memory (CD-ROM: Compact Disc Read-Only Memory), and the magnetic surface memory can be a magnetic disk memory or a magnetic tape memory. The volatile memory may be a random access memory (RAM: Random Access Memory) used as an external cache.By way of illustrative but non-limiting examples, many forms of RAM are available, such as static random access memory (SRAM), synchronous static random access memory (SSRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), sync link dynamic random access memory (SLDRAM), direct rambus random access memory (DRRAM), and the like. The memory described in the embodiments of the present disclosure is intended to include these and any other suitable types of memory, but is not limited thereto.

[0198] In an exemplary embodiment, the embodiments of the present disclosure further provide a storage medium, that is, a computer storage medium, specifically, a computer-readable storage medium, for example, including a memory in which a computer program is stored, and the computer program can perform the above steps of the method on the terminal side when executed by a first processor 72 of the terminal 70. The computer-readable storage medium may be a memory such as FRAM, ROM, PROM, EPROM, EEPROM, Flash Memory, magnetic surface memory, optical disk, or CD-ROM.

[0199] Note that terms such as "first" and "second" are for distinguishing similar objects and are not used to explain a specific order or sequence.

[0200] Furthermore, the technical solutions described in the embodiments of the present disclosure can be arbitrarily combined without conflict.

[0201] The above are only preferred embodiments of the present disclosure and are not intended to limit the protection scope of the present disclosure.

Claims

1. A channel state information (CSI) reporting method executed by a terminal, comprising: transmitting a CSI report to a network device, wherein the CSI report includes N sets of reporting amounts, and each set of reporting amounts includes an index of at least one reference signal and a layer 1 reference signal received power (L1-RSRP) or a layer 1 signal-to-interference-plus-noise ratio (L1-SINR) corresponding to the at least one reference signal; wherein N represents the number of sets of reporting amounts set by the network device, and N is a positive integer, the CSI reporting method.

2. Slots applied to at least one reference signal in each set of reporting amounts and the L1-RSRP or L1-SINR corresponding to the at least one reference signal are determined based on a first period set by the network device. The CSI reporting method according to claim 1.

3. The CSI reporting method further includes: quantizing the maximum L1-RSRP value or the maximum L1-SINR value in the N sets of reporting amounts with K bits, and using the maximum L1-RSRP value or the maximum L1-SINR value as a reference, calculating each L1-RSRP value or L1-SINR value other than the maximum L1-RSRP value or the maximum L1-SINR value in the N sets of reporting amounts as a difference to obtain a differential L1-RSRP value or a differential L1-SINR value, wherein the differential L1-RSRP value or the differential L1-SINR value is quantized with P bits; both K and P are positive integers, and K is greater than P. The CSI reporting method according to claim 1 or 2.

4. The CSI report further includes indication information, and the indication information indicates position information of the maximum L1-RSRP value or the maximum L1-SINR value in at least one L1-RSRP or L1-SINR included in the N sets of reporting amounts. The CSI reporting method according to claim 3.

5. The CSI reporting method further includes: performing non-uniform quantization on the L1-RSRP value or the L1-SINR value corresponding to each reference signal in the N sets of reporting amounts by a first artificial intelligence (AI) module. The CSI reporting method according to claim 1 or 2.

6. The total number of bits for performing non-uniform quantization on the L1-RSRP value or the L1-SINR value corresponding to each reference signal in the N sets of reporting amounts is smaller than a first threshold value. The CSI reporting method according to claim 5.

7. The CSI reporting method further includes: obtaining auxiliary information transmitted from the network device, where the auxiliary information includes channel state information and / or scheduling information; The CSI reporting method according to claim 5.

8. The CSI reporting method further includes: transmitting related information of the first AI module to the network device; The CSI reporting method according to claim 5.

9. The CSI reporting method further includes: when there are multiple reporting amount sets satisfying a predetermined condition among the M reporting amount sets, retaining one reporting amount set from the multiple reporting amount sets, excluding other reporting amount sets except the retained one reporting amount set from the M reporting amount sets to obtain N reporting amount sets, where M is a positive integer, M is equal to N, or M is greater than N; The CSI reporting method according to claim 2.

10. The CSI report further includes at least one reference signal index in the excluded other reporting amount sets, or The CSI report further includes L1-RSRP or L1-SINR corresponding to at least one reference signal in the excluded other reporting amount sets; The CSI reporting method according to claim 9.

11. Satisfying the predetermined condition includes: all or partially the same reference signal indexes for every two reporting amount sets among the multiple reporting amount sets, or for every two reporting amount sets among the multiple reporting amount sets, calculating the ratios of L1-RSRP or L1-SINR corresponding to each reference signal in one of the two reporting amount sets to L1-RSRP or L1-SINR corresponding to each reference signal in the other reporting amount set to obtain a plurality of ratios, and all the plurality of ratios are smaller than a second threshold, or for every two reporting amount sets among the multiple reporting amount sets, calculating the differences of L1-RSRP or L1-SINR corresponding to each reference signal in one of the two reporting amount sets to L1-RSRP or L1-SINR corresponding to each reference signal in the other reporting amount set to obtain a plurality of differences, and all the plurality of differences are smaller than a second threshold; The CSI reporting method according to claim 9.

12. A channel state information (CSI) reporting method executed by a network device, ​ Receiving a CSI report transmitted from a terminal, the CSI report including N sets of reporting amounts, each set of reporting amounts including an index of at least one reference signal and a layer 1 reference signal received power (L1-RSRP) or a layer 1 signal-to-interference-plus-noise ratio (L1-SINR) corresponding to the at least one reference signal, An N represents the number of sets of reporting amounts set by the network device, and N is a positive integer, a CSI reporting method.

13. Slots applied to at least one reference signal in each set of reporting amounts and the L1-RSRP or L1-SINR corresponding to the at least one reference signal are determined based on a first period set by the network device. The CSI reporting method according to claim 12.

14. The CSI reporting method further includes: further setting, for the terminal, the number N of sets of reporting amounts and the first period. The CSI reporting method according to claim 13.

15. The CSI report further includes indication information, and the indication information indicates position information of at least one L1-RSRP or L1-SINR included in N sets of reporting amounts of a maximum L1-RSRP value or a maximum L1-SINR value. The CSI reporting method according to claim 12.

16. The CSI reporting method further includes: restoring, by a second artificial intelligence (AI) module, the L1-RSRP or L1-SINR of each reference signal of N sets of non-uniformly quantized reporting amounts. The CSI reporting method according to claim 12.

17. The total number of bits for performing non-uniform quantization on the L1-RSRP value or L1-SINR value corresponding to each reference signal in N sets of reporting amounts is smaller than a first threshold value. The CSI reporting method according to claim 16.

18. The CSI reporting method further includes: further transmitting, to the terminal, auxiliary information, the auxiliary information including channel state information and / or scheduling information. The CSI reporting method according to claim 16.

19. The CSI reporting method further includes: further receiving, from the terminal, related information of the first AI module. The CSI reporting method according to claim 16.

20. When there are a plurality of sets of reporting amounts satisfying a predetermined condition among M sets of reporting amounts, holding one reporting quantity set from the plurality of reporting quantity sets, and further excluding other reporting quantity sets except the held one reporting quantity set from M reporting quantity sets to obtain N reporting quantity sets, where M is a positive integer, M is equal to N, or M is greater than N, The CSI reporting method according to claim 12.

21. The CSI report further includes indexes of at least one reference signal in the other reporting quantity sets that are excluded. or The CSI report further includes L1-RSRP or L1-SINR corresponding to at least one reference signal in the other reporting quantity sets that are excluded. The CSI reporting method according to claim 20.

22. Satisfying the predetermined condition means that the indexes of the reference signals for every two reporting quantity sets in the plurality of reporting quantity sets are all the same or partially the same, or for every two reporting quantity sets in the plurality of reporting quantity sets, calculating the ratios of L1-RSRP or L1-SINR corresponding to each reference signal in one of the two reporting quantity sets to L1-RSRP or L1-SINR corresponding to each reference signal in the other reporting quantity set to obtain a plurality of ratios, and all of the plurality of ratios are smaller than a second threshold value, or for every two reporting quantity sets in the plurality of reporting quantity sets, calculating the differences between L1-RSRP or L1-SINR corresponding to each reference signal in one of the two reporting quantity sets and L1-RSRP or L1-SINR corresponding to each reference signal in the other reporting quantity set to obtain a plurality of differences, and all of the plurality of differences are smaller than a second threshold value, The CSI reporting method according to claim 20.

23. A channel state information (CSI) reporting device, comprising a transmission unit configured to transmit a CSI report to a network device, the CSI report includes N reporting quantity sets, and each reporting quantity set includes indexes of at least one reference signal and layer 1 reference signal received power (L1-RSRP) or layer 1 signal-to-interference-plus-noise ratio (L1-SINR) corresponding to the at least one reference signal, where N represents the number of reporting quantity sets set by the network device, and N is a positive integer, the CSI reporting device.

24. A channel state information (CSI) reporting device, comprising A receiving unit configured to receive a CSI report transmitted from a terminal, the CSI report including N sets of reporting amounts, each set of reporting amounts including an index of at least one reference signal and layer 1 reference signal received power (L1-RSRP) or layer 1 signal-to-interference-plus-noise ratio (L1-SINR) corresponding to the at least one reference signal, An N represents the number of sets of reporting amounts set by the network device, and N is a positive integer, a CSI reporting device. **Claim 25** A terminal comprising a first processor and a first memory storing a computer program executable by the processor, The first processor executes all processes in the CSI reporting method according to any one of claims 1 to 11 by executing the computer program, a terminal. **Claim 26** A network device comprising a second processor and a second memory storing a computer program executable by the processor, The second processor executes all processes in the CSI reporting method according to any one of claims 12 to 22 by executing the computer program, a network device. **Claim 27** A computer-readable storage medium storing a computer program for causing a processor to execute all processes in the CSI reporting method according to any one of claims 1 to 11 or to implement all processes in the CSI reporting method according to any one of claims 12 to 22

Images