Communication method, related apparatus, and storage medium

Abstract

The present application relates to the technical field of wireless communications, and provides a communication method, a related apparatus, and a storage medium. The method is applied to a base station side, and comprises: sending resource configuration information to a user device (UE), the resource configuration information being used for indicating an uplink physical resource; and receiving monitoring data sent by the UE on the basis of the uplink physical resource, the monitoring data being used for monitoring a channel state information (CSI) prediction module and / or a two-sided model for CSI compression in the base station and / or a terminal device.

Description

Communication methods, related devices and storage media Technical Field

[0001] This application relates to the field of wireless communication technology, and in particular to a communication method, related apparatus and storage medium. Background Technology

[0002] Within the framework of Massive-MIMO communication systems, the measurement and feedback of Channel State Information (CSI) is one of the core technologies of the physical layer. Accurate measurement and feedback of CSI information can help the gNB implement correct data modulation and coding schemes, directly improving the spectral efficiency of the communication system.

[0003] 3GPP NR Rel-18 investigated CSI compression in both the spatial and frequency domains to increase CSI accuracy while reducing air interface overhead for CSI reporting. Figure 1-1 illustrates a flowchart of CSI spatial-frequency domain compression based on Artificial Intelligence (AI) / Machine-Learning (ML). This process is based on an encoder-decoder model. The encoder, deployed on the User Equipment (UE) side, uses spatial-frequency domain channel information (such as precoding matrices and eigenvectors) obtained from channel measurements as model input and outputs compressed channel information. The decoder, deployed on the base station side, uses the compressed channel information as model input and outputs complete channel information. Furthermore, a quantizer can be deployed after the encoder to further reduce the transmission overhead of compressed channel information, forming a complete CSI generation module. Correspondingly, a dequantizer needs to be deployed before the decoder to transform the quantized data into unquantized compressed channel information, forming a complete CSI reconstruction module. Furthermore, quantizers and dequantizers can also be integrated into encoders and decoders to reduce model complexity.

[0004] During the use of AI / ML models, it is necessary to monitor the performance of AI / ML models, including monitoring the CSI prediction module and the CSI compression two-sided model, in order to determine whether the current AI / ML model performance is good or meets the requirements. However, existing monitoring of CSI prediction compression still has a series of problems. Therefore, how to provide a more complete monitoring solution to improve the accuracy of model monitoring and ensure the normal operation of the model is an urgent issue to be considered. Summary of the Invention

[0005] This application provides a communication method, related device, and storage medium to improve the accuracy of model monitoring and ensure the normal operation of the model.

[0006] To achieve the above objectives, this application adopts the following technical solution:

[0007] The first aspect of this application provides a communication method applied to a base station, comprising: sending resource configuration information to a user equipment (UE), the resource configuration information being used to indicate uplink physical resources; and receiving monitoring data sent by the UE based on the uplink physical resources, the monitoring data being used to monitor the channel state information (CSI) prediction module and / or CSI compression two-sided model in the base station and / or the terminal device.

[0008] A second aspect of this application also provides a communication method applied to a UE side, comprising: determining uplink physical resources; and reporting monitoring data to a base station based on the uplink physical resources, wherein the monitoring data is used to monitor the CSI prediction module and / or CSI compression two-sided model in the base station and / or the terminal device.

[0009] A third aspect of this application also provides a communication method applied to a UE side, comprising: determining downlink physical resources for reconstructing CSI, so as to receive the reconstructed CSI; and associating the reconstructed CSI and monitoring data sent by the base station.

[0010] A fourth aspect of this application also provides a communication method applied to a base station, comprising: indicating configured downlink physical resources to a UE, and sending a reconfigured CSI based on the downlink physical resources.

[0011] A fifth aspect of this application also provides a wireless communication device, comprising: a processor and a memory, wherein the memory is used to store a computer program, and the processor is used to call and run the computer program stored in the memory to perform the method as described in any of the preceding embodiments.

[0012] A sixth aspect of this application also provides a computer-readable storage medium, the computer-readable storage medium including instructions that, when executed, cause the method described in any of the preceding claims to be implemented. Attached Figure Description

[0013] Figure 1-1 is a flowchart of a possible AI / ML-based CSI spatial frequency domain compression.

[0014] Figure 1-2 shows a possible trigger for data reporting.

[0015] Figure 2 is a flowchart illustrating a possible communication method provided in an embodiment of this application;

[0016] Figure 3 is a flowchart illustrating another possible communication method provided in an embodiment of this application;

[0017] Figure 4 is a schematic diagram of the storage of a possible wireless communication device provided in an embodiment of this application. Detailed Implementation

[0018] For ease of understanding, the relevant technologies involved in the embodiments of this application will be described below.

[0019] The technical solutions of the embodiments of this application will now be described with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0020] It should be understood that the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this article generally indicates that the preceding and following related objects have an "or" relationship.

[0021] To better understand this solution, this application will list the existing technologies that may be utilized in this solution. In the Rel-18 project, a codebook based on the spatiotemporal frequency domain was introduced for the Doppler-based CSI prediction method, so that the base station can obtain CSI information in a timely manner and adjust the data transmission mode when the UE is moving at high speed. The relevant technologies of this project include the following two points:

[0022] Related technical point 1: CSI calculation time

[0023] If Downlink Control Information (DCI) triggers the reporting of CSI on the Physical Uplink Shared Channel (PUSCH), the UE will report the nth CSI report triggered by the DCI if the following conditions are met:

[0024] Case 1: The first uplink symbol reported by the CSI triggered by this DCI is not earlier than symbol Z, provided that timing advance (TA) is included. ref ;

[0025] Case 2: The first uplink symbol carrying the nth CSI report is no earlier than symbol Z' if it includes TA. ref (n).

[0026] The above Z ref Defined as at least T after the last symbol of DCI proc,CSI =(Z)(2048+144)·κ2 -μ ·T C +T switch The first rising symbol, Z' ref (n) is defined as the periodic CSI-RS resource used for the most recent channel measurement corresponding to the nth CSI report, at least T' after the last symbol. proc,CSI =(Z')(2048+144)·κ2 -μ ·T C The first upward symbol.

[0027] In scenarios where an enhanced Type II codebook or an enhanced Type II port selection codebook is reported, the value of (Z,Z') is (Z2,Z′2), where Z′2 is the CSI calculation time, and the value is shown in Table 5.4-2 below.

[0028] In the Rel-18 CSI prediction scenario, i.e., codebookType IE is set to 'typeII-Doppler-r18' or 'typeII-Doppler-PortSelection-r18', the value of (Z,Z') can take the following four cases:

[0029] If N4 = 1 is configured, the CSI-RS resource set is aperiodic and each resource set is configured with K CSI-RS resources, then (Z, Z') takes the value (Z2 + 14(K-1)m, Z′2), where m is the time slot interval between adjacent CSI-RS resources;

[0030] If N4 = 1 is configured, the CSI-RS resource set is periodic or semi-persistent, and each resource set is configured with 1 CSI-RS resource, then (Z, Z') takes the value (Z2 + w, Z'2), where w = 56. (K P –1) or 56.K P A symbol, K P ∈{1,2,4} represents the number of active resources, indicated by UE capabilities;

[0031] If N4>1 is configured, the CSI-RS resource set is aperiodic and each resource set is configured with K CSI-RS resources, then (Z,Z') takes the value of (Z2+14(K-1)m,Z′2) or (Z2+14(K-1)m+Z′2,2Z′2), the value of which depends on the UE capability;

[0032] If N4>1 is configured, the CSI-RS resource set is periodic or semi-persistent and each resource set is configured with 1 CSI-RS resource, then (Z,Z') takes the value of (Z2+w,Z′2) or (Z2+w+Z′2,2Z′2), the value of which depends on the UE capability.

[0033] Table 5.4-2:CSI computation delay requirement 2

[0034] Table 5.4-2: CSI Calculation Delay Requirements 2

[0035] Related Technical Point 2: CSI Part 2 Parameter Priority

[0036] When a CSI report on the PUSCH consists of two parts, the UE may discard part 2 of the CSI. Part 2 CSI discarding follows a priority order according to existing protocols (such as TS38.214), as shown in Table 5.2.3-1 below, where N... Rep This is the number of CSI reports configured to be transmitted on the PUSCH. Priority 0 is the highest priority, Priority 2N... Rep It is the lowest priority, CSI report n is N Rep The nth smallest Pri in the CSI report iCSI The CSI report corresponds to the value (y,k,c,s). For CSI report n, the subbands indicated by the value of 1 in the higher-layer parameter csi-ReportingBand are numbered sequentially in ascending order, with the lowest subband designated as subband 0. When a Part 2 CSI of a specific priority is discarded, the UE discards all information bits of that priority.

[0037] For Enhanced Type II reports and Enhanced Type II for predicted PMI with high-level parameter configuration of N4=1, i is indexed by l,i,f. 2,4,l i 2,5,l i 1,7,l Each element is associated with a priority value Pri(l,i,f) = 2·L·υ·π(f) + υ·i + l, where f = 0, 1, ..., M υ -1 indicates that the smaller the priority value, the higher the priority of the element. In this codebook type, Group 0 contains i 1,1 i 1,2 i 1,8,l (l=1,…,υ), Group 1 contains i 1,5 i1,6,l i 1,7,l middle The highest priority element, i 2,3,l i 2,4,l middle The highest priority element, i 2,5,l middle The elements with the highest priority (l = 1, ..., υ). Group 2 contains i 1,7,l middle The element with the lowest priority, i 2,4,l middle The element with the lowest priority, i 2,5,l middle The element with the lowest priority (l=1,…,υ).

[0038] For Enhanced Type II for predicted PMI with high-level parameter configuration of N4>1, i indexed by l,i,f 2,4,l i 2,5,l i 1,7,l Each element is associated with a priority value Pri(l,i,f,j)=2L·υ·M ν ·j+2L·υ·f+υ·i+l, where l=1,2,…,υ, i=0,1,…,2L-1, f=0,1,…,M υ -1, j = 0, 1, the smaller the priority value, the higher the priority of the element. Under this type of codebook, Group 0 contains i 1,1 i 1,2 i 1,8,l (l=1,…,υ), the second broadband CQI, Group 1 contains i 1,5 i 1,6,l i 1,7,l middle The highest priority element, i 2,3,l i 2,4,l middle The highest priority element, i 2,5,l middle The highest priority element, i 1,10,l (l=1,…,υ), the second subband CQI of the even-numbered subbands. Group 2 contains i 1,7,l middle The element with the lowest priority, i 2,4,l min(K) NZ - The element with the lowest priority, i 2,5,l middle The element with the lowest priority (l=1,…,υ), the second subband CQI of the odd-numbered subbands.

[0039] Table 5.2.3-1:Priority reporting levels for Part 2CSI

[0040] Table 5.2.3-1: Reporting Priority of Part 2 CSI

[0041] In addition, for clarity, the monitoring data related to the CSI prediction compression network-side model monitoring involved in this application's technical solution falls into four categories, which will be explained below:

[0042] Predicted CSI: The CSI output after inputting historical measurement channels into the CSI prediction module;

[0043] Compressed CSI: The CSI output after the prediction channel is input into the encoder and quantizer;

[0044] Measured CSI: Measure the channel output CSI at the time point in which the predicted CSI is located;

[0045] Intermediate monitoring data: Monitoring information output by the UE-side auxiliary monitoring system, including monitoring results and / or monitoring indicator values.

[0046] Monitoring of CSI prediction compression models is divided into network-side monitoring and UE-side monitoring. Monitoring based on prediction CSI corresponds to monitoring of the CSI compression two-sided model. Monitoring based on measurement CSI can be further divided into monitoring prediction CSI as input to the CSI generation module and monitoring measurement CSI as input to the CSI generation module. Monitoring prediction CSI as input to the CSI generation module corresponds to end-to-end monitoring, including monitoring of both the CSI prediction module and the CSI compression two-sided model. Monitoring measurement CSI as input to the CSI generation module corresponds to monitoring of the CSI compression two-sided model. Existing schemes for monitoring CSI prediction compression models have a series of problems, including the following:

[0047] For network-side model monitoring of CSI prediction compression, since a solution for independent monitoring of the CSI prediction module has not yet been discussed, when monitoring decisions are made on the network side, the network side needs to collect monitoring data. One solution is for the network side to monitor the CSI prediction module using the measured CSI and predicted CSI reported by the UE. This solution does not rely on the UE's ability to calculate model monitoring indicators, allowing for more autonomous and flexible network-side decision-making, but it consumes more uplink resources. Another solution is for the UE to perform auxiliary monitoring, i.e., after completing the monitoring of the CSI prediction module, the UE reports relatively low-overhead intermediate monitoring data to the network side to reduce uplink resource consumption. However, this solution requires specific monitoring indicators and corresponding UE capabilities to avoid the network side not supporting the output results of some monitoring methods used by different UE vendors, thus affecting the judgment of model performance. Therefore, the network side needs to configure the required monitoring data to the UE based on UE capabilities, communication resources, and monitoring requirements. There are four possible combinations of reported monitoring data:

[0048] Combination 1: Compressed CSI + Measuring CSI: Network-side monitoring of end-to-end model performance, joint monitoring of the CSI prediction module and the CSI compression dual-side model;

[0049] Combination 2: Compressed CSI + Measured CSI + Predicted CSI: The network side simultaneously and independently monitors the performance of both the CSI prediction module and the CSI compression dual-side model;

[0050] Combination 3: Compressed CSI + Predicted CSI + Monitoring Intermediate Quantities: The network side monitors the performance of the CSI compression model, with the monitoring label being predicted CSI, while the UE assists in monitoring the performance of the CSI prediction module;

[0051] Combination 4: Compressed CSI + Measured CSI + Monitoring Intermediate Quantities: The network side monitors the performance of the CSI compression model, with the monitoring tag being measured CSI. At the same time, the UE assists in monitoring the performance of the CSI prediction module.

[0052] According to the existing CSI reporting mechanism, the network side can trigger multiple aperiodic CSI reports based on PUSCH or multiple semi-persistent CSI reports based on PUSCH through DCI. Therefore, when performing CSI prediction compression model monitoring on the network side, the reporting of the above monitoring data can be triggered through DCI, as shown in Figure 1-2, which is a possible triggering data reporting diagram provided by the embodiment of this application. In this way, since each triggered CSI report configuration corresponds to the same CSI-RS resource set, the following problem exists: The UE cannot determine the PUSCH time-frequency resources for each monitoring data report, thus affecting the network side's acquisition of the required monitoring data.

[0053] Furthermore, no solutions have yet been discussed for UE-side model monitoring of CSI prediction compression. If the UE-side model monitoring solution used in CSI compression is still adopted, where the base station uses the reconstructed CSI precoding reference signal, end-to-end model monitoring is only possible, and independent monitoring of the CSI prediction module and the CSI compression dual-side model is not possible. A feasible and less costly approach is for the base station to send the reconstructed CSI to the UE, allowing the UE to monitor both the CSI prediction module and the CSI compression dual-side model. This raises issue 2: how to correlate the reconstructed CSI with the measurement CSI and prediction CSI output internally by the UE; otherwise, the false alarm rate of model monitoring will increase.

[0054] Furthermore, this application also notes that regardless of whether the CSI prediction compression model is monitored on the network side or the UE side, there are some common problems, such as:

[0055] In Rel-18 CSI prediction, the parameters related to CSI calculation time consider the duration of Doppler-based CSI prediction and the solution time of the space-time-frequency basis precoding matrix indicator (PMI) compared to non-CSI prediction (i.e., traditional CSI estimation). In AI / ML-based CSI prediction compression, the computational complexity of the CSI compression bilateral model and the AI ​​model that may be applied in the CSI prediction module is affected by various factors, such as model format and model structure. Compared to Rel-18 channel prediction and PMI calculation, CSI prediction compression is more flexible in terms of CSI calculation time. In addition, the introduction of quantization / dequantization modules will also increase the CSI calculation time. Therefore, problem 3 exists: the calculation time of UE-side compressed CSI will be different from the calculation time of Rel-18 CSI prediction.

[0056] Furthermore, in traditional mechanisms, UEs are typically configured to report multiple CSIs for different functions, such as link adaptation, beam management, and time-frequency tracking. The introduction of CSIs for model monitoring further strains already limited uplink physical resources, and the UE may need to report CSIs for model monitoring to multiple serving cells. This leads to questions 4 and 5.

[0057] Question 4: The transmission resources for multiple CSI reports are prone to overlap, leading to a decrease in system performance. In the existing CSI conflict handling rules, the UE determines the priority of each CSI report based on the CSI reporting configuration ID, serving cell index, CSI reporting content, and CSI reporting method, further determining whether to reuse or discard the CSI. The existing CSI conflict handling rules affect the data integrity of CSIs used for model monitoring. For example, if one type of monitoring data or non-monitoring data is reported using aperiodic PUSCH, and another type of monitoring data is reported using semi-persistent PUSCH, the monitoring data reported using semi-persistent PUSCH will be discarded when it overlaps with aperiodic PUSCH due to its lower priority.

[0058] Question 5: Multiple monitoring data can be transmitted via a single PUSCH, but with limited resources, it cannot be guaranteed that all data will be transmitted through this PUSCH. In the existing PUSCH-based CSI reporting mechanism, CSI is split into CSI part 1 and CSI part 2. CSI part 2 further reports based on CSI reporting priority and CSI parameter priority, and discards some parameters. Because compressed CSI differs from the content of traditional Type I / Type II PMI, the existing mechanism is no longer suitable for PUSCH transmission carrying compressed CSI.

[0059] Furthermore, during network-side monitoring, measured CSI and predicted CSI are reported using PMI. During UE-side monitoring, reconstructed CSI is also sent to the UE using PMI. Ideally, during model monitoring, raw monitoring data should be used to ensure the accuracy of model monitoring. However, PMI introduces errors when representing the precoding matrix, leading to problem 6: these errors cause a decrease in the accuracy of model monitoring.

[0060] In view of this, the embodiments of this application provide a series of solutions to various problems, which will be described separately in two cases: 1. Model monitoring is on the base station side; 2. Model monitoring is on the UE side. Specifically, these include the following:

[0061] I. Please refer to Figure 2, which is a flowchart of a possible communication method provided in an embodiment of this application, including at least one of the following steps:

[0062] 201. The base station sends monitoring data reporting configuration to the UE;

[0063] The base station sends a monitoring data reporting configuration to the UE. This configuration indicates the requested monitoring data and its reporting format to support UE-side auxiliary model monitoring. The monitoring data reporting configuration includes, but is not limited to, at least one of the following parameters: report quantity, compressed CSI auxiliary information, compressed CSI auxiliary information granularity, monitoring metrics, monitoring metric value dimension, and / or monitoring metric threshold. The indication methods for each parameter include, but are not limited to, at least one of the following: direct indication, index indication, single-bit indication, and / or bitmap indication. Each parameter will be described separately below.

[0064] Optional indication methods for reported volume:

[0065] 1) Direct indication: Optional reporting quantities include uncompressed CSI, compressed CSI, UE-side monitoring indicator values, and UE-side monitoring results. The descriptions of each reporting quantity are as follows:

[0066] If the reported quantity is uncompressed CSI, the UE reports at least one of the following parameters obtained from measuring CSI-RS: CSI-RS Resource Indicator (CRI), Rank Indicator (RI), Layer Indicator (LI), PMI, and / or Channel Quality Indicator (CQI), wherein the PMI is reported according to the codebook configuration indicated by the base station.

[0067] If the indicated reporting quantity is compressed CSI, the UE reports at least one of the following parameters: CRI, RI, LI, compressed precoding matrix, and / or CQI, wherein the compressed precoding matrix is ​​the data output after inputting the historical measurement channel into the CSI prediction module and the CSI generation module, and the base station can input the compressed precoding matrix into the base station-side CSI reconstruction module to obtain the reconstructed precoding matrix.

[0068] If the indicated reporting quantity is a monitoring indicator value on the UE side, the UE reports the monitoring indicator value calculated by predicting CSI and measuring CSI. The monitoring indicator is configured by the base station, and the base station can use an internal evaluation algorithm to judge the performance of the CSI prediction module based on the monitoring indicator value.

[0069] If the indicated reporting quantity is the monitoring result on the UE side, then the UE reports a judgment on the performance of the CSI prediction module after comparing the monitoring indicator values ​​with the monitoring indicator thresholds indicated by the base station. For example, 0 indicates that the CSI prediction module performance is abnormal, and 1 indicates that the CSI prediction module performance is normal. The base station can directly know the performance of the CSI prediction module through this monitoring result and make subsequent decisions. The UE uses at least one of the following rules to judge the performance of the CSI prediction module: i. If K monitoring indicator values ​​are not less than the corresponding thresholds, then the CSI prediction module performance is normal; otherwise, the CSI prediction module performance is abnormal. The number K≥1 is indicated by the base station. When K is equal to the number of monitoring indicator values, it means that the CSI prediction module performance is normal when all monitoring indicator values ​​are not less than the corresponding thresholds. For example, this rule can be used when the monitoring indicator is Squared Generalized Cosine Similarity (SGCS); ii. If K monitoring indicator values ​​are not greater than the corresponding threshold, the CSI prediction module performs normally; otherwise, the CSI prediction module performs abnormally. The number K ≥ 1 is indicated by the base station. When K equals the number of monitoring indicator values, it means that the CSI prediction module performs normally when all monitoring indicator values ​​are not greater than the corresponding threshold. For example, this rule can be used when the monitoring indicator is Normalized Mean Squared Error (NMSE).

[0070] 2) A bitmap indicator is used, with the number of bits corresponding to the number of reporting types. This number uses a default value and / or is determined by the UE's capabilities. Each bit corresponds to one reporting type. Specifically, when a bit is 1, it instructs the UE to report the reporting type corresponding to that bit; otherwise, it does not report.

[0071] 3) An index is used as an indicator. Each index corresponds to a reporting quantity. At least one bit can be used to represent the index. The number of bits uses a default integer value and / or is determined by the UE capability.

[0072] Optional indication methods for compressing CSI auxiliary information:

[0073] 1) Direct indication. Optional auxiliary information includes quantization error of compressed precoding matrix and codebook adjustment value. Quantization error represents the difference between each element of the compressed precoding matrix before and after quantization. It is applicable to scalar quantization (SQ) and vector quantization (VQ). The base station can obtain more accurate input data for the CSI reconstruction module based on the compressed precoding matrix and its quantization error reported by the UE. Codebook adjustment value represents the difference between each element of the quantized codebook before and after fine-tuning. It is applicable to VQ. The base station can update the quantized codebook based on the codebook adjustment value to improve quantization accuracy.

[0074] 2) A bitmap is used, with the number of bits corresponding to the number of auxiliary information types. This number uses a default value and / or is determined by the UE's capabilities. Each bit corresponds to one type of auxiliary information. Specifically, when a bit is 1, it instructs the UE to report the auxiliary information corresponding to that bit; otherwise, it does not report it.

[0075] 3) An index is used as an indicator. Each index corresponds to a type of auxiliary information. At least one bit can be used to represent the index. The number of bits uses a default integer value and / or is determined by the UE capability.

[0076] It should be noted that if this parameter, i.e., compressed CSI auxiliary information, is defaulted, it means that the amount of information reported by the base station is not compressed CSI or the UE does not need to report compressed CSI auxiliary information.

[0077] Optional indication methods for compressing CSI auxiliary information granularity:

[0078] 1) Direct indication, the optional auxiliary information granularity includes integer and floating-point numbers, where the information carried by the integer indication is represented by SQ quantization, and the information carried by the floating-point indication does not need to be quantized. The number of bits corresponding to the information granularity can be a predefined value and / or indicated by higher-level parameters.

[0079] 2) A single bit is used for indication, with 0 and 1 each indicating a granularity of auxiliary information. For example, 0 indicates an integer and 1 indicates a floating-point number.

[0080] Optional indicator display methods for monitoring metrics:

[0081] It should be noted that this parameter exists when the base station indicates that the reported quantity is a UE-side monitoring indicator value or a UE-side monitoring result.

[0082] 1) Direct indication; optional monitoring metrics include, but are not limited to, SGCS and NMSE.

[0083] 2) A bitmap indicator is used, with the number of bits corresponding to the number of monitoring metrics. This number uses a default value and / or is determined by the UE's capabilities. Each bit corresponds to one monitoring metric. If a bit is 1, it instructs the UE to use the monitoring metric corresponding to that bit to calculate the monitoring metric value; otherwise, it does not. An example is a monitoring metric that includes SGCS and NMSE. The bitmap has 2 bits. Assuming the first bit represents SGCS and the second bit represents NMSE, then a bitmap of 10 bits instructs the UE to calculate SGCS based on the predicted CSI and measured CSI.

[0084] 3) Index indicators are used, with each index corresponding to a monitoring metric. An index can be represented by at least one bit, with the number of bits using a default integer value and / or determined by the UE's capabilities. An example is a 1-bit index where index values ​​0 and 1 represent SGCS and NMSE, respectively.

[0085] Optional indication methods for monitoring indicator value dimensions:

[0086] It should be noted that this parameter indicates the calculation dimension of the monitoring indicator value.

[0087] 1) A bitmap is used for indication. The number of bits corresponds to the number of dimensions of the precoding matrix / tensor. This number uses a default value and / or is determined by the UE capability. Each bit corresponds to one matrix dimension. If a bit is 1, the UE is instructed to calculate the monitoring index value for each value of the dimension corresponding to that bit. If multiple bits are 1, the UE is instructed to calculate the monitoring index value for each value of multiple dimensions corresponding to multiple bits. If all bits are 0, the UE is instructed to calculate the monitoring index value based on all matrix elements. An example is a precoding matrix / tensor whose dimensions include prediction time point, subband, antenna port, and layer. The bitmap has 4 bits. Assuming the first bit to the last bit represent the prediction time point, subband, antenna port, and layer respectively, a bitmap of 1000 indicates that the UE calculates the monitoring index value for each precoding matrix / tensor element corresponding to each prediction time point, and a bitmap of 1100 indicates that the UE calculates the monitoring index value for each subband corresponding to each prediction time point.

[0088] 2) An index indicator is used, in which one index value indicates that the UE calculates the monitoring indicator value based on all matrix elements, and the remaining index values ​​correspond to a combination of precoded matrix / tensor dimensions. The index can be represented by at least one bit, and the number of bits uses a default integer value and / or is determined by the UE capability.

[0089] Optional indication methods for monitoring indicator thresholds:

[0090] It should be noted that this parameter exists when the base station indicates that the reported quantity is the monitoring result on the UE side. The UE compares the calculated monitoring indicator value with the monitoring indicator threshold to determine the performance of the CSI prediction module. The base station can indicate at least one monitoring indicator threshold according to the granularity of the monitoring indicator.

[0091] 1) Direct indication; optional thresholds may include {0, 0.1, 0.2, ..., 0.8, 0.9, 1}.

[0092] 2) An index is used as an indicator, with each index corresponding to a threshold. At least one bit can be used to represent the index, and the number of bits uses a default integer value. An example is a 4-bit index, where the index values ​​0 to 11 represent {0, 0.1, 0.2, ..., 0.8, 0.9, 1} respectively.

[0093] 202. The base station sends resource configuration information to the UE;

[0094] In addition, the base station also schedules uplink physical resources to carry monitoring data to ensure that the base station can successfully receive the required model monitoring data. The base station sends resource configuration information to the UE, which is used to indicate the uplink physical resources. Among them, the indication method of the uplink physical resources corresponding to each monitoring data in the resource configuration information includes, but is not limited to, at least one of the following methods: indicating by configuring the reporting configuration for each monitoring data type, indicating by configuring the monitoring data type corresponding to the CSI reporting configuration information associated with the CSI semi-persistent reporting trigger state and / or the CSI aperiodic reporting trigger state, and / or indicating the reporting order of monitoring data types through downlink control information (DCI). The following will explain each indication method.

[0095] i. Instructions are given by configuring monitoring data reporting settings for each type of monitoring data;

[0096] The base station is configured with at least one of the following monitoring data reporting configurations, enabling the UE to determine the physical resources corresponding to each monitoring data based on the monitoring data type corresponding to the CSI reporting configuration triggered by DCI and the indicated physical resource time-frequency allocation. As mentioned above, the monitoring data types include: compressed CSI, predicted CSI, measured CSI, and monitoring intermediate quantities. The reporting configurations for each monitoring data type are as follows:

[0097] At least one compressed CSI reporting configuration, i.e., the CSI reporting configuration indicates that the monitored data type is compressed CSI and / or the CSI reporting configuration is dedicated to compressed CSI;

[0098] At least one predictive CSI reporting configuration, i.e., a CSI reporting configuration that indicates the monitoring data type is predictive CSI and / or a CSI reporting configuration dedicated to predictive CSI;

[0099] At least one measurement CSI reporting configuration, i.e., the CSI reporting configuration indicates that the monitoring data type is measurement CSI and / or the CSI reporting configuration is dedicated to measurement CSI;

[0100] At least one monitoring intermediate quantity reporting configuration, i.e., the CSI reporting configuration indicates that the monitoring data type is monitoring intermediate quantity and / or the CSI reporting configuration is dedicated to monitoring intermediate quantity.

[0101] For ease of understanding, an example of a base station configuring CSI and reporting the configuration to the UE via Radio Resource Control (RRC) signaling can be described as follows:

[0102] ii. Indicate by configuring the monitoring data type corresponding to the CSI reporting configuration information (CSI-AssociatedReportConfigInfo IE) associated with the CSI semi-persistent reporting trigger state (CSI-SemiPersistentOnPUSCH-TriggerState IE) and / or the CSI aperiodic reporting trigger state;

[0103] That is, the base station configures the corresponding monitoring data type for each CSI semi-persistent reporting trigger state and / or CSI aperiodic reporting trigger state associated with the CSI reporting configuration information, so that the UE can determine the physical resource corresponding to the monitoring data type according to the configured monitoring data type and the corresponding triggered CSI reporting configuration.

[0104] Based on the above description, an example of the base station configuring the CSI aperiodic reporting trigger state to the UE via RRC signaling is as follows:

[0105] iii. Use DCI to indicate the reporting order of monitored data types.

[0106] The DCI that triggers CSI reporting contains a field indicating the reporting order of monitored data types. This field uses an index to indicate the physical resources corresponding to each monitored data type. Each index corresponds to a set of monitoring data types and can be represented by at least one bit. The number of bits uses a default integer value and / or is determined by the number of reported monitoring data types. The above reporting order is defined as at least one of the following:

[0107] 1. If the reporting order of a certain type of monitoring data is earlier than that of another type of monitoring data, then the first symbol of the physical resource where the monitoring data is located is earlier than the first symbol of the physical resource where the other monitoring data is located. If the first symbols are the same, then the frequency of the first physical resource block (PRB) or resource element (RE) of the physical resource where the monitoring data is located is lower than that of the first PRB or RE of the physical resource where the other monitoring data is located.

[0108] 2. If the reporting order of a certain monitoring data type is earlier than that of another monitoring data type, then the frequency of the first PRB or RE of the physical resource where the monitoring data is located is lower than that of the first PRB or RE of the physical resource where the other monitoring data is located. If the frequencies of the first PRB or RE are the same, then the first symbol of the physical resource where the monitoring data is located is earlier than the first symbol of the physical resource where the other monitoring data is located.

[0109] For example, when the number of monitored data types is 3, the index table corresponding to the field of the reporting order of monitored data types can be as shown in Table 1. Among them, monitored data type A, monitored data type B, and monitored data type C can all take any of the following data types: predicted CSI, compressed CSI, measured CSI, or monitored intermediate quantity.

[0110] Table 1 Index table when the number of monitored data types is 3

[0111] Alternatively, when the number of monitored data types is 2, the index table corresponding to the field of the reporting order of monitored data types can be as shown in Table 2. Both monitored data type A and monitored data type B can take any of the following data types: predicted CSI, compressed CSI, measured CSI, or monitored intermediate quantity.

[0112] Table 2 Index table when the number of monitored data types is 2

[0113] It should be noted that in this embodiment, the monitoring data reporting configuration sent by the base station to the UE in step 201 and the resource configuration information sent by the base station to the UE in step 202 can be integrated into one message or sent separately. When sent separately, step 201 can be executed first and then step 202, or step 202 can be executed first and then step 201, or they can be executed simultaneously. That is, the timing of steps 201 and 202 is not limited in this application.

[0114] 203. The UE calculates the monitoring data configured by the base station;

[0115] The UE obtains historical channel information by measuring CSI-RS and calculates the monitoring data configured by the base station based on this historical channel information. This historical channel information includes, but is not limited to, the channel matrix and / or precoding matrix. When the monitoring data includes the calculation time of CSI, this application provides several schemes for determining the CSI calculation time, including but not limited to: 1. Determining based on the calculation time of predicted CSI; 2. Depends on at least one of the following parameters: CSI prediction module calculation time, encoder calculation time, and / or quantizer calculation time; 3. Based on a predefined calculation time, which will be explained separately below.

[0116] 1. Determined based on the calculation time of the predicted CSI;

[0117] Let the calculation time for predicting CSI be Z′. preCSI The definition of this can be found in the relevant technical point one mentioned above, and will not be repeated here. The computation time Z' for compressing CSI is then calculated. compCSI It may be in at least one of the following forms:

[0118] Compressed CSI calculation time is the predicted CSI calculation time plus an offset, i.e., Z'. compCSI =Z' preCSI +Δ Z′,1 , where Δ Z′,1 To calculate the time offset, in symbols, the value is determined using at least one of the following methods: UE capability, base station indication, and / or a predefined value;

[0119] or,

[0120] The compressed CSI calculation time is the rounded value of the product of the predicted CSI calculation time and the offset factor. or Where k1 is the offset factor, which is a positive number and is determined by at least one of the following methods: UE capability, base station indication, and / or a predefined value.

[0121] 2. It depends on at least one of the following parameters: CSI prediction module calculation time, encoder calculation time, and / or quantizer calculation time. Each parameter will be described below.

[0122] 1) CSI prediction module calculation time

[0123] When the CSI prediction module uses a non-AI algorithm, the CSI prediction module's computation time Z' pre It may be in at least one of the following forms:

[0124] a. The CSI prediction module calculation time is the predicted CSI calculation time minus the offset, i.e., Z' pre =Z' preCSI -Δ Z′,2 , where Δ Z′,2 To calculate the time offset, this value is a non-negative real number in sign, and is determined using at least one of the following methods: UE capability, base station indication, and / or a predefined value;

[0125] b. The CSI prediction module calculation time is the rounded value of the product of the predicted CSI calculation time and the offset factor, i.e. or Where k2 is the offset factor, which is a positive number not greater than 1, and is determined by at least one of the following methods: UE capability, base station indication, and / or using a predefined value;

[0126] c. The CSI prediction module calculates the time using a predefined value in symbols. This predefined value is a default integer value and / or depends on at least one of the following parameters: the non-AI algorithm used and / or the subcarrier spacing.

[0127] For example, non-AI algorithms include at least one of the following: Moving Average (MA), Exponential Smoothing (ES), and / or Autoregressive (AR) models, in which case the CSI prediction module calculates the time Z'. pre It can be determined according to the index in Table 3.

[0128] Table 3. Calculation time Z' of CSI prediction module based on non-AI algorithm pre

[0129] When the CSI prediction module uses an AI model, the CSI prediction module's computation time Z' pre Predefined values ​​are used, which depend on at least one of the following parameters: model type, number of model layers / modules, number of model parameters, number of model floating point operations (FLOPs), and / or subcarrier spacing, where the number of model layers / modules is a parameter value or parameter range, the number of model parameters is a parameter value or parameter range, and FLOPs is a parameter value or parameter range.

[0130] Optionally, the computation time of the CSI prediction module can be determined by at least one of the following parameters: model type, number of model layers / modules, and / or subcarrier spacing, wherein the model type includes at least one of the following: Fully Connected Neural Network (FCN), Convolutional Neural Network (CNN), Recurrent Neural Network (RNN), and / or Transformer model, and the number of model layers / modules is expressed as a parameter range, thereby determining the computation time Z′ of the CSI prediction module. pre The index can be determined from Table 4, which is for a specific subcarrier interval, and each subcarrier interval has a corresponding index table.

[0131] Table 4. Calculation time Z' of the CSI prediction module based on the AI ​​model pre

[0132] Alternatively, the calculation time of the CSI prediction module may be determined by FLOPs and / or subcarrier spacing, where FLOPs are in the form of a parameter range, and thus the calculation time Z′ of the CSI prediction module is determined. pre It can be determined according to the index in Table 5.

[0133] Table 5. Calculation time Z′ of the CSI prediction module based on the AI ​​model. pre

[0134] 2) Encoder calculation time

[0135] Encoder calculation time Z' code Predefined values ​​are used, in symbol form. These predefined values ​​are default integer values ​​and / or depend on at least one of the following parameters: model type, number of model layers / modules, number of model parameters, FLOPs, number of models, and / or subcarrier spacing. The number of model layers / modules is a parameter value or range, the number of model parameters is a parameter value or range, FLOPs is a parameter value or range, and the number of models is determined based on the model's functionality. If the model infers independently for each layer, the number of models is greater than 1 when the rank is greater than 1.

[0136] For example, the encoder computation time is determined by FLOPs and the number of models. FLOPs are expressed in the form of parameter ranges. When the model infers independently for each layer and the rank is greater than 1, there are two cases:

[0137] Case 1: The model is a layer-common model. In this case, the encoder computation time can be expressed as Z′. code =I×P FLOPs ,I∈{1,…,P NUM}

[0138] In the above formula, P NUM Indicates the number of models, where I is the number of model groups. Groups can be divided based on cores or threads, determined by the UE's capabilities. P FLOPs This indicates the model computation time corresponding to the FLOPs range of the shared model in this layer. This value can be determined based on the table index.

[0139] Case 2: The model is a layer-specific model. In this case, the encoder computation time can be expressed as:

[0140] In the above formula, I represents the number of model groups. Groups can be divided based on cores or threads, determined by the UE's capabilities. This indicates the j-th element in the i-th group. i The computation time of a model within the FLOPs range can be determined using the index in Table 6.

[0141] Table 6. Computation time P for shared / specific models at each layer FLOPs

[0142] 3) Quantizer calculation time

[0143] Quantizer calculation time Z′ quant A predefined value is used, in symbols, which is a default integer value and / or depends on at least one of the following parameters: the quantization algorithm used, and / or the subcarrier spacing. For example, the quantization algorithm includes SQ and / or VQ, and the quantizer calculation time Z' is... quant It can be determined according to the index in Table 7.

[0144] Table 7 Quantizer Calculation Time Z' quant

[0145] Compress CSI computation time Z' compCSI Determined by at least one of the following parameters: CSI prediction module calculation time Z' pre Encoder calculation time Z' code And / or quantizer computation time Z' quant It can be represented as: Z′ compCSI =f(Z' pre ,Z′ code ,Z′ quant )

[0146] Optionally, the computation time for compressed CSI can be determined jointly by the computation time of the CSI prediction module, the encoder computation time, and the quantizer computation time, i.e., Z' compCSI =Z' pre +Z' code +Z' quant .

[0147] Optionally, given that the encoder includes quantizer functionality to form a complete CSI generation module, or that the quantizer has low complexity and its computation time is much shorter than that of the CSI prediction module and the encoder, the computation time for compressed CSI is determined jointly by the computation time of the CSI prediction module and the encoder, i.e., Z'. compCSI =Z′ pre +Z' code .

[0148] Optionally, if the CSI prediction module uses a non-AI algorithm with low complexity, its computation time is much shorter than that of the quantizer and / or encoder. In this case, the computation time for compressed CSI is determined by the combined computation time of the encoder and quantizer, i.e., Z'. compCSI =Z′ code +Z' quant Or it can be determined solely by the encoder's calculation time, i.e., Z' compCSI =Z′ code .

[0149] 3. Based on predefined computation time.

[0150] The computation time for compressed CSI is calculated using a predefined value, which is N times the CSI computation time for a specific codebook type. Here, N is contained in the set {1, 2, 3, 4, 6}, meaning the predefined value includes Z′2, 2Z′2, 3Z′2, 4Z′2, and / or 6Z′2. The specific codebook type includes, but is not limited to, at least one of the following types: Type II codebook, Type II port selection codebook, enhanced Type II codebook, enhanced Type II port selection codebook, and / or further enhanced Type II port selection codebook. This application does not impose any specific limitations on the type.

[0151] 204. UE determines uplink physical resources;

[0152] The UE determines the uplink physical resources to report monitoring data based on those resources. In this embodiment, the UE can determine the uplink physical resources in various ways, including but not limited to: 1. determining the uplink physical resources according to base station instructions; 2. determining the physical resources corresponding to each monitoring data according to a predefined reporting order of monitoring data types; 3. determining the uplink physical resources based on the CSI calculation time. These will be described separately below:

[0153] 1. Determine uplink physical resources based on base station instructions.

[0154] Referring to step 201, the base station sends resource configuration information to the UE. This resource configuration information indicates uplink physical resources, enabling the UE to determine the uplink physical resources based on the base station's instructions. Further details are omitted here.

[0155] 2. Determine the physical resources corresponding to each monitoring data according to the predefined reporting order of the monitoring data types. The meaning of the reporting order of the monitoring data types is as described in step 201, and will not be repeated here.

[0156] The predefined reporting order for monitoring data types can be at least one of the following:

[0157] i. Monitoring data type A appears before monitoring data type B, and monitoring data type B appears before monitoring data type C;

[0158] ii. Monitoring data type A appears before monitoring data type C, and monitoring data type C appears before monitoring data type B;

[0159] iii. Monitoring data type B appears before monitoring data type A, and monitoring data type A appears before monitoring data type C;

[0160] iv. Monitoring data type B appears before monitoring data type C, and monitoring data type C appears before monitoring data type A;

[0161] v. Monitoring data type C precedes monitoring data type A, and monitoring data type A precedes monitoring data type B;

[0162] vi. Monitoring data type C precedes monitoring data type B, and monitoring data type B precedes monitoring data type A.

[0163] It is worth noting that the monitoring data types A, B, and C mentioned above can all be any of the following data types: predicted CSI, compressed CSI, measured CSI, or monitoring intermediate quantities.

[0164] 3. Determine the uplink physical resources based on the CSI calculation time.

[0165] Define the latency Z corresponding to each monitored data type. delay =Δ offset +Z', where Δ offset Z′ represents the symbol interval between the last symbol of the DCI that triggered the CSI report and the first or last symbol of the most recently measured CSI-RS resource. When the monitored data type is predictive CSI or compressed CSI, the most recently measured CSI-RS resource is the historical CSI-RS resource of the last measurement. When the monitored data type is measured CSI or monitored intermediate quantity, the most recently measured CSI-RS resource is the last CSI-RS resource corresponding to the predictive CSI. Z′ represents the calculation time of CSI. The calculation time definitions for predictive CSI, measured CSI, and monitored intermediate quantity are given in the above-mentioned related technical point one. The calculation time definition for compressed CSI can be referred to step 202 of this embodiment, and will not be repeated here. The UE determines the delay time Z′ according to the data type of the monitored data. delay The size determines the reporting order of each monitored data type. For example, the latency Z corresponding to each monitored data type... delay The smaller the value, the earlier it is reported.

[0166] Optionally, the UE can also determine the reporting order of each monitoring data type based on the CSI calculation time Z' corresponding to the monitoring data type. The calculation times for predicted CSI, measured CSI, monitoring intermediate quantities, and compressed CSI are described above and will not be repeated here. For example, the smaller the CSI calculation time Z' corresponding to the monitoring data type, the earlier its reporting order.

[0167] 205. The UE reports monitoring data to the base station based on uplink physical resources;

[0168] After determining the uplink physical resources, the UE reports the calculated monitoring data through those uplink physical resources. Considering that the UE may also report non-monitoring data based on uplink physical resources, when the resources for monitoring data or the resources for monitoring data and non-monitoring data (CSI) overlap in the time domain, the reporting of monitoring data can be determined according to priority rules. These priority rules can be shared by monitoring data and non-monitoring data (CSI) or dedicated to monitoring data. The following will describe these two scenarios separately.

[0169] Specifically, in this embodiment of the application, when the priority rule is that monitoring data and non-monitoring data CSI are shared, if the physical resources where the two monitoring data transmissions are located have at least one symbol overlap or the physical resources where the monitoring data and non-monitoring data CSI are located have at least one symbol overlap, then the reporting method needs to be determined according to the CSI priority. The priority of CSI reporting depends on at least one of the following parameters: reporting configuration ID, serving cell index, reporting content, reporting type, and / or the type of monitoring data carried.

[0170] If at least one type of monitoring data reporting has a higher priority than non-monitoring data CSI reporting, then the priority value of CSI reporting can be expressed as follows: Pri CSI (f,y,k,c,s)=8·N cells ·M s ·f+2·N cells ·M s ·y+N cells ·M s ·k+M s • c + s; Formula 1

[0171] In Formula 1, f represents the monitoring data type when CSI reports monitoring data; y=0 indicates aperiodic CSI reporting based on PUSCH; y=1 indicates semi-persistent CSI reporting based on PUSCH; y=2 indicates semi-persistent CSI reporting based on PUCCH; y=3 indicates periodic CSI reporting based on PUCCH; k=0 indicates CSI reporting carrying L1-RSRP or L1-SINR; k=1 indicates CSI reporting not carrying L1-RSRP or L1-SINR; c represents the serving cell index; s represents the reporting configuration ID; and N... cells M represents the maximum number of serving cells. s This indicates the maximum number of CSI reporting configuration IDs. The lower the priority value for a CSI report, the higher its priority.

[0172] Based on Formula 1, for example, compressed CSI reports have a higher priority than CSI reports that do not carry compressed CSI. In this case, f=0 in Formula 1 represents compressed CSI reports, and f=1 represents CSI reports that do not carry compressed CSI.

[0173] Based on Formula 1, for example, the priority of compressed CSI reporting and measurement CSI reporting is higher than that of CSI reporting without compressed CSI or measurement CSI. In this case, f=0 in Formula 1 represents compressed CSI reporting or measurement CSI reporting, and f=1 represents CSI reporting without compressed CSI or measurement CSI. Furthermore, if the priority of compressed CSI reporting is higher than that of measurement CSI reporting, then f=0 in Formula 1 represents compressed CSI reporting, f=1 represents measurement CSI reporting, and f=2 represents CSI reporting without compressed CSI or measurement CSI.

[0174] Based on Formula 1, for example, the priority of reported monitoring data is higher than that of reported non-monitoring data (CSI). In this case, f=0 in Formula 1 represents reported monitoring data, and f=1 represents reported non-monitoring data (CSI). Further, if the priority of compressed CSI reporting is higher than that of other monitoring data (including measurement CSI, predicted CSI, and monitoring intermediate quantities), then f=0 in Formula 1 represents compressed CSI reporting, f=1 represents other monitoring data reporting, and f=2 represents reported non-monitoring data (CSI). Even further, if the priority of measurement CSI reporting is higher than that of predicted CSI reporting or monitoring intermediate quantity reporting, then f=0 in Formula 1 represents compressed CSI reporting, f=1 represents measurement CSI reporting, f=2 represents predicted CSI reporting or monitoring intermediate quantity reporting, and f=3 represents reported non-monitoring data (CSI).

[0175] Additionally, when priority rules are applied specifically to monitoring data, if the physical resources where two monitoring data transmissions reside overlap by at least one symbol, the reporting method can be determined based on the priority of the monitoring data transmissions. The priority of monitoring data transmissions depends on, but is not limited to, at least one of the following parameters: reporting configuration ID, serving cell index, monitoring data type, and / or data volume, as specifically defined below:

[0176] CSI reported configuration ID: The value range is {0,1,…,M} s -1}, where M s The maximum number of configurations that can be reported to CSI;

[0177] Serving Cell Index: The serving cell receiving CSI reports is represented by an index, with values ​​ranging from {0, 1, ..., M}. c -1}, where M c To the maximum number of serviced communities;

[0178] Monitoring data types include: compressed CSI, predicted CSI, measured CSI, and monitoring intermediate quantities;

[0179] Data volume: Represented by indexes, each index corresponds to a range of monitoring data volume, and the index value range is {0,1,…,M}. d -1}, where M d This is the maximum index value.

[0180] Optionally, an example of a monitoring data transmission priority rule is as follows: Compressed CSI has a higher priority than other monitoring data (including measured CSI, predicted CSI, and monitoring intermediate quantities); when the priority of monitoring data types is the same, the larger the data volume, the higher the priority of monitoring data transmission; when the data volume is the same, the smaller the serving cell index, the higher the priority of monitoring data transmission; when the serving cell index is the same, the smaller the CSI reporting configuration ID, the higher the priority of monitoring data transmission. Therefore, the priority formula associated with monitoring data transmission is: Pri monitor (f,d,c,s)=M d ·M c ·M s ·f+M c ·M s ·(M d -1-d)+M s Formula 2 (c+s)

[0181] In Formula 2, f=0 indicates compressed CSI reporting, f=1 indicates reporting of other monitoring data, d indicates data volume index, it can be assumed that the smaller the index value, the smaller the data volume, and vice versa, c indicates serving cell index, and s indicates CSI reporting configuration ID.

[0182] Furthermore, if the priority of compressed CSI is greater than that of measured CSI, and the priority of measured CSI is greater than that of predicted CSI or monitoring intermediate quantities, then in Formula 2, f=0 indicates compressed CSI reporting, f=1 indicates measured CSI reporting, and f=2 indicates predicted CSI reporting or monitoring intermediate quantity reporting.

[0183] Optionally, in this application embodiment, another example of the monitoring data transmission priority rule is as follows: To ensure the stability of the primary serving cell's CSI prediction compression monitoring, the smaller the serving cell index, the higher the priority of monitoring data transmission. When the serving cell index is the same, compressed CSI has a higher priority than other monitoring data (including measured CSI, predicted CSI, and monitoring intermediate quantities). When the monitoring data type priorities are the same, the larger the data volume, the higher the priority of monitoring data transmission. When the data volume is the same, the smaller the CSI reporting configuration ID, the higher the priority of monitoring data transmission. Therefore, the priority formula associated with monitoring data transmission is: Pri monitor (f,d,c,s)=2·M d ·M s·c+M d ·M s ·f+M s ·(M d Formula 3 (-1-d)+s

[0184] In Formula 3, f=0 indicates compressed CSI reporting, f=1 indicates reporting of other monitoring data, d indicates data volume index, it is assumed here that the smaller the index value, the smaller the data volume, and vice versa, c indicates serving cell index, and s indicates CSI reporting configuration ID.

[0185] Furthermore, if the priority of compressed CSI is higher than that of measured CSI, and the priority of measured CSI is higher than that of predicted CSI or monitored intermediate quantities, then the priority formula for monitoring data transmission is Pri. monitor (f,d,c,s)=3·M d ·M s ·c+M d ·M s ·f+M s ·(M d Formula 4 (-1-d)+s

[0186] In Formula 4, f=0 indicates compressed CSI reporting, f=1 indicates measured CSI reporting, and f=2 indicates predicted CSI reporting or monitoring intermediate quantity reporting.

[0187] Optionally, in this embodiment, another example of the monitoring data transmission priority rule is: compressed CSI has a higher priority than other monitoring data (including measurement CSI, predicted CSI, and monitoring intermediate quantities). When the monitoring data types have the same priority, the smaller the CSI reporting configuration ID, the higher the priority of the monitoring data transmission. Therefore, the priority formula associated with monitoring data transmission is Pri. monitor (f,s)=M s Formula 5 (f+s)

[0188] In Formula 5, f=0 indicates compressed CSI reporting, f=1 indicates reporting of other monitoring data, and s indicates the CSI reporting configuration ID.

[0189] Furthermore, if the priority of compressed CSI is greater than the priority of measured CSI, and the priority of measured CSI is greater than the priority of predicted CSI or monitoring intermediate quantities, then in the above formula, f=0 indicates compressed CSI reporting, f=1 indicates measured CSI reporting, and f=2 indicates predicted CSI reporting or monitoring intermediate quantity reporting.

[0190] In summary, the lower the priority value calculated by the above priority value formula for monitoring data transmission, the higher the priority of the monitoring data transmission, and vice versa.

[0191] In addition, in this embodiment of the application, when the first condition is met, for example, when multiple monitoring data reports are carried on a single PUSCH, the reporting of monitoring data can be further divided into a first part (Part 1) of data and a second part (Part 2) of data. The reporting of Part 2 data follows a priority rule, which depends on the reporting priority of the monitoring data and the group to which the data belongs. Specifically, Part 1 of predicted CSI and measured CSI includes, but is not limited to, at least one of the following parameters: RI, complete CQI or the first CQI, and the number of non-zero amplitude coefficients between layers; Part 1 of compressed CSI includes, but is not limited to, at least one of the following parameters: the number of elements in the compressed precoding matrix, the layer indicator carried, and / or the subspace indicator carried, which will be described separately below:

[0192] RI: This parameter indicates the number of layers;

[0193] Number of elements in the compressed precoding matrix: If the model infers independently for each layer, this parameter indicates the number of elements in a single-layer compressed precoding matrix; otherwise, it indicates the number of elements in all layers of the compressed precoding matrix. This parameter can be used to determine the information bits of Group 0, Group 1, and Group 2 in Part 2CSI when the quantizer algorithm is SQ, and to determine the information bits of Group 1 and Group 2 in Part 2CSI when the quantizer algorithm is VQ.

[0194] Layer Indicator: Indicates the layer number corresponding to the information carried by Group 1 and Group 2. It can be represented by a bitmap, with the number of bits being the maximum number of layers. Each bit in the bitmap corresponds to one layer. If a bit is 1, it indicates that Group 1 and Group 2 of Part 2 CSI carry the information of that layer. This parameter is defaulted if the compressed model infers all layers simultaneously. This parameter can be used to indicate layers with large quantization errors or large codebook adjustments to save reporting overhead. The determination of the magnitude of quantization error or codebook adjustment is based on the UE implementation.

[0195] Subspace Indicator: Indicates the subspace ordinal number corresponding to the information carried by Group 1 and Group 2. This can be represented by a bitmap, with the number of bits equal to the maximum number of subspaces. Each bit in the bitmap corresponds to one subspace. If a bit is 1, it indicates that Group 1 and Group 2 of Part 2 CSI carry that subspace information. This parameter is defaulted if SQ quantization or VQ quantization is used and subspace division is not implemented. This parameter can be used to indicate subspaces with larger codebook adjustments to save reporting overhead. The determination of the codebook adjustment size is based on the UE implementation.

[0196] For Part 2, both the predicted CSI and the measured CSI consist of a second CQI and / or PMI, while Part 2 CSI for compressed CSI includes, but is not limited to, at least one of the following parameters: compressed precoding matrix, quantization error, and / or codebook adjustment value. Part 2 CSI of the monitoring data is discarded according to the priority order shown in Table 8 below, where N... Rep This configures the number of monitoring data reports transmitted on the PUSCH. Priority 0 is the highest priority, and Priority 2N is the lowest. Rep It is the lowest priority; the nth monitoring data report is N. Rep The monitoring data report corresponding to the nth lowest priority value in the monitoring data report. For the priority of monitoring data reports, please refer to step 205. The details will not be repeated here.

[0197] Table 8 Monitoring Data Part 2: CSI Priority

[0198] In addition, for the information contained in Group 0, Group 1, and Group 2 above regarding predicted CSI and measured CSI, please refer to the relevant technical point 2 above, and will not be repeated here.

[0199] For compressed CSI, assuming the number of layers is L and the number of VQ partition subspaces is S, Group 0 in Table 8 above includes, but is not limited to, at least one of the following information:

[0200] 1) The compressed precoding matrix after SQ quantization, which exists in at least one of the following forms: if the compressed model infers simultaneously on all layers, then the information is a single SQ quantized compressed precoding matrix; and / or, if the compressed model infers separately on each layer (L≥2), then the information is L SQ quantized compressed single-layer precoding matrices.

[0201] 2) The compressed precoding matrix after VQ quantization exists in at least one of the following forms:

[0202] i. If the compressed model infers simultaneously across all layers and the VQ does not partition into subspaces, then the information is a single VQ codebook index, and each VQ codebook index corresponds to a compressed precoding matrix;

[0203] ii. If the compressed model infers simultaneously on all layers and the VQ is divided into subspaces, then the information consists of S VQ codebook indices, each VQ codebook index corresponding to a compressed precoding matrix in its subspace;

[0204] iii. If the compressed model infers separately for each layer (L≥2) and the VQ does not divide the subspace, then the information consists of L VQ codebook indices, each VQ codebook index corresponding to a compressed single-layer precoding matrix;

[0205] iv. If the compressed model infers separately for each layer (L≥2) and the VQ is divided into subspaces, then the information consists of SL VQ codebook indices, each VQ codebook index corresponding to a compressed single-layer precoding matrix within its subspace.

[0206] The information bits contained in Group 0 depend on at least one of the following parameters: quantization algorithm, quantization data type, and / or the number of compressed precoding matrix elements, indicated by higher-level parameters and / or Part 1CSI.

[0207] For compressed CSI, if Group 1 and Group 2 carry information, assuming the number of layers carried is L and the number of VQ subspaces carried is S, then Group 1 and Group 2 contain at least one of the following combinations of information:

[0208] A. If the compressed model infers on all layers simultaneously, Group 1 contains the quantization error of each element of the compressed precoding matrix, while Group 2 does not contain information.

[0209] B. If the compressed model infers separately for each layer (L≥2), there exists at least one combination of information contained in Group 1 and Group 2:

[0210] Group 1 contains the quantization error of each element in L compressed single-layer precoding matrices; Group 2 contains no information.

[0211] Group 2: Group 1 includes the preceding The quantization error of each element in a compressed single-layer precoding matrix, Group 2 including... The quantization error of each element in a compressed single-layer precoding matrix;

[0212] Group 3: Group 1 includes the preceding... The quantization error of each element in a compressed single-layer precoding matrix, Group 2 including... The quantization error of each element in a compressed single-layer precoding matrix;

[0213] Group 4: Group 1 contains the quantization error of each element of a compressed single-layer precoding matrix with odd-numbered layers, and Group 2 contains the quantization error of each element of a compressed single-layer precoding matrix with even-numbered layers;

[0214] Group 5: Group 1 contains the quantization error of each element of a compressed single-layer precoding matrix with even-numbered layers, and Group 2 contains the quantization error of each element of a compressed single-layer precoding matrix with odd-numbered layers.

[0215] C. If VQ is used for quantization, and the VQ codebook is the same for each layer and for each subspace, then there exists at least one combination of information contained in Group 1 and Group 2:

[0216] Group 1: Group 1 contains codebook adjustment values ​​for S subspaces; Group 2 contains no information.

[0217] Group 2: Group 1 includes the preceding The codebook adjustment value for each subspace, after Group 2 includes it. Codebook adjustment value for each subspace (S≥2);

[0218] Group 3: Group 1 includes the preceding... The codebook adjustment value for each subspace, after Group 2 includes it. Codebook adjustment value for each subspace (S≥2);

[0219] Group 4: Group 1 contains codebook adjustment values ​​for subspaces with odd ordinal numbers, and Group 2 contains codebook adjustment values ​​for subspaces with even ordinal numbers;

[0220] Group 5: Group 1 contains codebook adjustment values ​​for subspaces with even ordinal numbers, and Group 2 contains codebook adjustment values ​​for subspaces with odd ordinal numbers.

[0221] D. If VQ is used for quantization, and the VQ codebook used for each subspace and each layer is different, then there exists at least one combination of information contained in Group 1 and Group 2:

[0222] Group 1: Group 1 contains codebook adjustment values ​​for L layers and S subspaces, while Group 2 contains no information (L≥2);

[0223] Group 2: Group 1 includes the L-level front The codebook adjustment value for each subspace, after Group 2 includes L layers. Codebook adjustment values ​​for each subspace (L≥2, S≥2);

[0224] Group 3: Group 1 includes the L-level front The codebook adjustment value for each subspace, after Group 2 includes L layers. Codebook adjustment values ​​for each subspace (L≥2, S≥2);

[0225] Group 4: Group 1 contains codebook adjustment values ​​for subspaces with odd ordinal numbers of L levels, and Group 2 contains codebook adjustment values ​​for subspaces with even ordinal numbers of L levels (L≥2).

[0226] Group 5: Group 1 contains codebook adjustment values ​​for subspaces with even ordinal numbers of L levels, and Group 2 contains codebook adjustment values ​​for subspaces with odd ordinal numbers of L levels (L≥2).

[0227] The information bits contained in Group 1 and Group 2 depend on at least one of the following parameters: quantization algorithm, number of compressed precoding matrix elements, layer indicator of the bearer, subspace indicator of the bearer, and / or granularity of the bearer information, indicated by higher-level parameters and / or Part 1 CSI.

[0228] 206. The quality of the monitoring data sent by the UE to the base station;

[0229] In addition, to ensure the accuracy of model monitoring, the UE may optionally report the quality of each monitoring data to the base station. This quality may use, but is not limited to, at least one of the following parameters: the error between the precoding matrix / tensor obtained from PMI and the actual precoding matrix / tensor, candidate values ​​within a predefined range, and / or UE measurement values.

[0230] Specifically, the quality of CSI measurement and prediction can be expressed in at least one of the following ways:

[0231] 1) The error between the precoding matrix / tensor obtained from PMI and the actual precoding matrix / tensor can be expressed in at least one of the following forms:

[0232] i) Error based on at least one of the following dimensions: prediction time point, sub-band, antenna port, and / or, layer.

[0233] If based on one of the above dimensions, the UE calculates the error for each value of that dimension. For example, the error based on the prediction time point dimension means that the error is calculated for each prediction time point.

[0234] If based on the above multiple dimensions, the UE calculates the error for each value in the multiple dimensions. For example, the error based on the prediction time point and the sub-band dimension means that the error is calculated for each sub-band at each prediction time point.

[0235] ii) The error between all elements of the precoding matrix / tensor obtained from PMI and all elements of the actual precoding matrix / tensor;

[0236] The error indices used in the above error calculation include at least one of the following: SGCS, NMSE, and / or average error;

[0237] iii) After comparing each element of the precoding matrix / tensor obtained from PMI with that of the actual precoding matrix / tensor, output a quality indicator matrix / tensor and transform this quality indicator matrix / tensor into PMI form. The definition of each element in the quality indicator matrix / tensor is as follows: if the difference |xx′| between the precoding matrix / tensor element x' obtained from PMI and the actual precoding matrix / tensor element x is not greater than a threshold, then the quality indicator matrix / tensor element takes the value of the actual precoding matrix / tensor element; if the difference |xx′| is greater than the threshold, then the quality indicator matrix / tensor element takes the value of the actual precoding matrix / tensor element with a phase shift, for example, a phase shift of 90° or 180°. In this case, the quality indicator matrix / tensor element takes the value of... or e jπ •x, where the above thresholds use predefined values ​​and / or are indicated by higher-level parameters;

[0238] 2) Candidate values ​​within a predefined range. For example, {0, 0.1, 0.2, ..., 0.9, 1} is used as an indicator, where smaller numbers indicate worse CSI quality, and vice versa.

[0239] In addition, in this embodiment, step 205 is for the UE to report monitoring data to the base station, and step 206 is for the UE to report the quality of each monitoring data to the base station. The monitoring data and the quality of each monitoring data can be sent as a single message or sent separately. When sent separately, step 205 can be executed first and then step 206, or step 206 can be executed first and then step 205, or they can be executed simultaneously. Therefore, the timing of the two steps is not limited in this application.

[0240] 207. The base station performs model monitoring based on the monitoring data and executes model / function decisions.

[0241] After receiving monitoring data sent by the UE based on physical uplink resources, the base station performs model monitoring based on the monitoring data and executes model / function decisions.

[0242] Optionally, if the UE also reports the quality of the monitoring data, the base station can also filter the monitoring data according to the CSI quality involved in step 206 above and use it for model monitoring or use the CSI quality to compensate for the monitoring results in order to improve the accuracy of model monitoring. The methods for filtering the monitoring data and compensating for the monitoring results are implemented on the base station side.

[0243] Given that base stations perform model monitoring and execute model / function decisions based on monitoring data, which is a relatively mature technology, this application may adopt existing technology solutions or possible future solutions, which will not be elaborated here.

[0244] In summary, this embodiment addresses and explains the various problems existing in the prior art. Specifically, for problem 1, a monitoring data reporting format configuration and a matching mechanism between monitoring data and configured physical resources are proposed. On the one hand, this enables the system to support UE-side auxiliary monitoring, reducing uplink resource consumption; on the other hand, it supports the network side in collecting the required monitoring data and making monitoring decisions. Regarding problem 3, step 203 in this embodiment proposes to compress the CSI calculation time, solving the problem that the Rel-18 prediction CSI calculation time is not suitable because it does not consider AI and quantization operations, thus improving the flexibility of compressing CSI calculation time and supporting UE decisions regarding monitoring data reporting. For problems 4 and 5, step 204 in this embodiment proposes a model monitoring data reporting priority rule, solving the problem of uplink physical resource consumption. The problem of overlapping CSI transmission resources under limited source conditions is addressed by ensuring the integrity of monitoring data. Furthermore, the mechanism of multiple monitoring data reuse PUSCH proposed in step 205 solves the problem of insufficient uplink physical resources for complete reporting of monitoring data through scheduled PUSCH, thereby improving spectrum efficiency and ensuring the accuracy of CSI reconstruction. Regarding problem 6, step 206 in this embodiment proposes that the UE also uploads monitoring data of quality, i.e., CSI quality. This solves the problem of decreased model monitoring accuracy caused by errors introduced by PMI when representing the precoding matrix. This allows the UE to assist the network side in filtering monitoring data or compensating for monitoring results based on data quality, thereby improving the accuracy of model monitoring.

[0245] The communication method for model monitoring on the base station side has been described above. The following will describe the communication method for model monitoring on the UE side. II. Please refer to Figure 3, a flowchart of another possible communication method provided in this embodiment, including at least one of the following steps:

[0246] 301. The UE obtains historical channel information by measuring CSI-RS and outputs compressed CSI based on the historical channel information;

[0247] The UE obtains historical channel information by measuring CSI-RS and performs CSI prediction and compression output based on this historical channel information. In this embodiment, the calculation time for compressed CSI is determined by at least one of the following methods: 1. Based on the calculation time of predicted CSI; 2. Depends on at least one of the following parameters: CSI prediction module calculation time, encoder calculation time, and / or quantizer calculation time; 3. Based on a predefined calculation time. The specific determination method is similar to the method for determining the calculation time of compressed CSI in step 203 of the embodiment shown in Figure 2, and will not be elaborated further here.

[0248] 302. The UE reports compressed CSI to the base station;

[0249] After the UE outputs the compressed CSI, it reports the compressed CSI to the base station. If the compressed CSI overlaps in the time domain with the resources of monitoring data or with the resources of non-monitoring data CSI, the reporting of the compressed CSI is determined according to a priority rule. This priority rule can be shared by both monitoring data and non-monitoring data CSI. In this case, the CSI reporting priority depends on at least one of the following parameters: reporting configuration ID, serving cell index, reporting content, reporting type, and / or the type of monitoring data carried. Alternatively, the priority rule can be dedicated to monitoring data. In this case, the monitoring data reporting priority depends on at least one of the following parameters: reporting configuration ID, serving cell index, monitoring data type, and / or data volume.

[0250] In addition, if multiple monitoring data reports containing compressed CSI are carried on a single PUSCH, the reporting of compressed CSI can also be divided into Part 1 and Part 2. The reporting priority of Part 2 data is related to the reporting priority of compressed CSI and the data grouping within Part 2 data. Part 1 includes, but is not limited to, at least one of the following parameters: RI, number of compressed precoding matrix elements, layer indicator of the carrier, subspace indicator of the carrier, etc. Part 2 includes, but is not limited to, at least one of the following parameters: compressed precoding matrix, quantization error, and / or codebook adjustment value.

[0251] In this embodiment, the method by which the UE reports compressed CSI is similar to the method of compressed CSI in the monitoring data reported by the UE to the base station in step 205 of the embodiment shown in Figure 2, and will not be described in detail here.

[0252] 303. The base station reconstructs the CSI based on the compressed CSI output;

[0253] After receiving the compressed CSI sent by the UE, the base station inputs the compressed CSI into the CSI reconstruction module for base station-side model inference and outputs the reconstructed CSI (Recovery CSI). In this embodiment, the monitoring data related to the CSI prediction and compressed UE-side model monitoring is the reconstructed CSI.

[0254] 304. The base station configures the downlink physical resources for reconstructing the CSI to the UE and sends the reconstructed CSI;

[0255] Optionally, the base station configures the downlink physical resources for CSI reconfiguration to the UE, including: the base station sends the downlink physical resource allocation for CSI reconfiguration to the UE;

[0256] Alternatively, the base station can also indicate the configured downlink physical resources to the UE by configuring time-frequency related parameters. Specifically, the time-frequency related parameters include, but are not limited to, at least one of the following parameters:

[0257] 1) The offset of the time slot where the physical resource is located relative to the time slot where the physical resource carrying the compressed CSI is located: the unit is time slot, and the offset value can be a default integer value and / or configured by at least one of the following signaling: DCI, MAC CE, and / or RRC;

[0258] 2) The starting symbol of the physical resource in its time slot: This parameter uses a default integer value and / or is configured by at least one of the following signaling: DCI, Media Access Control (MAC) Control Element (CE), and / or RRC;

[0259] 3) Offset of the physical resource start symbol relative to the physical resource start symbol or last symbol carrying the compressed CSI: in symbols, the offset value may be a default integer value and / or configured by at least one of the following signaling: DCI, MAC CE, and / or RRC;

[0260] 4) The offset of the starting resource block of the physical resource relative to the starting resource block of the physical resource carrying the compressed CSI: in units of RB, the offset value can be a default integer value and / or configured by at least one of the following signaling: DCI, MAC CE, and / or RRC;

[0261] 5) Number of symbols occupied by physical resources: This parameter uses a default integer value and / or is configured by at least one of the following signaling: DCI, MAC CE, and / or RRC;

[0262] 6) Number of resource blocks occupied by physical resources: This parameter uses a default integer value and / or is configured by at least one of the following signaling: DCI, MAC CE, and / or RRC.

[0263] It should be noted that if the base station indicates at least one of the above time-frequency related parameters through DCI, the indication can be made in the DCI that triggers the compressed CSI reporting.

[0264] After configuring downlink physical resources, the base station sends a reconfiguration CSI to the UE based on those downlink physical resources. It should be noted that in this embodiment, the base station's configuration of downlink physical resources is optional; that is, in practical applications, the base station can also send the reconfiguration CSI to the UE based on predefined downlink resources.

[0265] 305. The quality of the reconstructed CSI sent by the base station to the UE;

[0266] Optionally, to ensure the accuracy of model monitoring, the base station may also send the quality of the reconstructed CSI to the UE. The quality of the reconstructed CSI may employ, but is not limited to, at least one of the following parameters: the error between the precoding matrix / tensor obtained from PMI and the actual precoding matrix / tensor, and / or candidate values ​​within a predefined range. The forms of these parameters are similar to those of the corresponding parameters involved in step 206 in the embodiment shown in Figure 2, and will not be elaborated further here.

[0267] In addition, in this embodiment of the application, in step 305, the base station sends the reconstructed CSI to the UE. Step 305 is optional, whereby the base station sends the quality of the reconstructed CSI to the UE. The reconstructed CSI and the quality of the reconstructed CSI can be sent as a single message or sent separately. When sent separately, the reconstructed CSI can be sent first and then the quality of the reconstructed CSI can be sent, or the quality of the reconstructed CSI can be sent first and then the reconstructed CSI can be executed, or they can be executed simultaneously. Therefore, this application does not limit the timing of the transmission.

[0268] 306. The UE determines the downlink physical resources and correlates and reconstructs the CSI with other monitoring data;

[0269] The UE determines downlink physical resources to receive the reconfiguration CSI sent by the base station based on those downlink physical resources. In this embodiment, the UE determines downlink physical resources in various ways, including but not limited to the following:

[0270] 1. The physical resources for bearer reconfiguration CSI are determined by the downlink physical resource allocation configured by the base station. That is, the base station configures the time and frequency resources for bearer reconfiguration CSI to the UE and instructs the UE accordingly.

[0271] 2. The physical resources for bearer reconfiguration CSI are determined by time-frequency related parameters. That is, the UE determines the physical resources for bearer reconfiguration CSI based on these time-frequency related parameters, which can be configured or predefined by the base station.

[0272] Therefore, after the UE determines the downlink physical resources carrying the reconstructed CSI, it receives the reconstructed CSI based on those downlink physical resources and associates it with other monitoring data, including compressed CSI and internally output measured and predicted CSI. Correspondingly, the UE can associate the reconstructed CSI with monitoring data in various ways, including but not limited to the following:

[0273] 1. The UE receives the downlink physical resource allocation configured by the base station and also receives association parameters sent by the base station. These association parameters are used by the UE to associate reconstructed CSI and monitoring data. The association parameters include, but are not limited to, at least one of the following: associated CSI-RS resource set index, associated CSI-RS resource index, associated compressed CSI reporting configuration ID, and / or compressed CSI reporting ID, to instruct the UE to associate the reconstructed CSI with the monitoring data. The monitoring data includes compressed CSI and the measured CSI and predicted CSI output internally by the UE. A detailed explanation of each association parameter is as follows:

[0274] Associated CSI-RS resource set index: The UE will use other monitoring data of CSI-RS output within the CSI-RS resource set that is closest in time interval to the time interval before the reconstructed CSI reception timing for CSI prediction compression as associated data;

[0275] Associated CSI-RS resource index: The UE will use other monitoring data from the CSI-RS output corresponding to the CSI-RS resource index that is closest in time interval to the time interval before the reconstructed CSI reception timing as associated data;

[0276] Associated Compressed CSI Reporting Configuration ID: The UE will use the compressed CSI corresponding to the reporting configuration with the closest time interval before the reconstructed CSI reception time, as well as the predicted CSI and measured CSI output through the same historical CSI-RS, as associated data;

[0277] Compressed CSI Report ID: Each compressed CSI report is represented by a compressed CSI report ID. The UE will associate the compressed CSI corresponding to the compressed CSI report ID with the closest time interval before the reconstructed CSI reception time, along with the predicted CSI and measured CSI output from the same historical CSI-RS, as related data. The compressed CSI report ID is determined through at least one of the following methods:

[0278] Method 1: The UE determines the compressed CSI reporting ID and reports the corresponding reporting ID each time a compressed CSI is reported;

[0279] Method 2: The UE determines the initial compressed CSI reporting ID. Compressed CSI reports triggered by the same DCI are accumulated based on this reporting ID and the reporting sequence number. For example, assuming the initial reporting ID is x, the ID corresponding to the first compressed CSI report triggered by the DCI is x, the ID corresponding to the second compressed CSI report is x+1, and so on.

[0280] Method 3: Predefine an initial compressed CSI reporting ID. Compressed CSI reports triggered by the same DCI are based on this reporting ID and accumulated according to the reporting sequence number. This initial reporting ID is a default integer value.

[0281] Method 4: The base station instructs the corresponding reporting ID to be reported each time the CSI is compressed;

[0282] Method 5: The base station indicates the initial compressed CSI reporting ID. Compressed CSI reports triggered by the same DCI are accumulated based on the reporting ID according to the reporting sequence number.

[0283] It should be noted that the downlink physical resource allocation and association parameters received by the UE can be sent by the base station through a single message, or the base station can send them separately through different messages; the specifics are not limited here.

[0284] 2. The UE reconstructs CSI and monitoring data based on time-frequency related parameters. In other words, time-frequency related parameters can be used to reconstruct CSI and compressed CSI, and further, they can be used to reconstruct CSI and the measurement CSI and prediction CSI output by the UE corresponding to the compressed CSI.

[0285] 307. The UE performs model monitoring and executes model / function decisions based on the monitoring data.

[0286] Optionally, if the base station also sends the quality of the reconstructed CSI to the UE, the UE can also filter the monitoring data based on the quality of the reconstructed CSI involved in step 305 above and use it for model monitoring, or use the quality of the reconstructed CSI to compensate for the monitoring results, so as to improve the accuracy of model monitoring. The methods for filtering the monitoring data and compensating for the monitoring results are implemented on the UE side.

[0287] In addition, given that the UE performs model monitoring and executes model / function decisions based on monitoring data, which is relatively mature in the existing technology, this application may adopt the existing technology solution or the possible solution in the future, which will not be elaborated here.

[0288] In summary, this embodiment addresses and explains problems 2 and others in the prior art. Specifically, for problem 2, a method for reconstructing the association between CSI and UE-side monitoring data is proposed, which reduces the physical resource consumption during UE-side model monitoring and ensures the accuracy of model monitoring.

[0289] The figures above illustrate in detail the communication method provided in the embodiments of this application. Please refer to Figure 4, which is a storage diagram of a wireless communication device in an embodiment of this application. The wireless communication device includes a processor and a memory. The memory stores computer programs, and the processor calls and runs the computer programs stored in the memory to execute the methods provided by any embodiment of the information processing method of this application and any non-conflicting combination thereof. The storage medium 20 of the wireless communication device in this embodiment stores instruction / program data 21. When the instruction / program data 21 is executed, it implements the methods provided by any embodiment of the information processing method of this application and any non-conflicting combination thereof. The instruction / program data 21 can be formed into a program file and stored in the storage medium 20 in the form of a software product, so that a computer device (which may be a personal computer, server, or base station, etc.) or processor executes all or part of the steps of the methods in various embodiments of this application. The aforementioned storage medium 20 includes various media capable of storing program code, such as a USB flash drive, mobile hard drive, read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk, or terminal devices such as computers, servers, mobile phones, and tablets.

[0290] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces, or indirect coupling or communication connection between apparatuses or units, and may be electrical, mechanical, or other forms.

[0291] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0292] The above are merely embodiments of this application and do not limit the scope of this patent application. Any equivalent structural or procedural changes made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of this application.

[0293] The above embodiments can be implemented, in whole or in part, by software, hardware (such as circuits), firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. A semiconductor medium can be a solid-state drive.

[0294] It should be understood that the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. A and B can be singular or plural. Additionally, the character " / " in this article generally indicates an "or" relationship between the preceding and following related objects, but it can also represent an "and / or" relationship. Please refer to the context for a more accurate understanding.

[0295] In this application, "at least one" means one or more, and "more than one" means two or more. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of single or multiple items. For example, at least one of a, b, or c can mean: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or multiple.

[0296] It should be understood that in the various embodiments of this application, the order of the above-mentioned processes does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0297] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0298] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0299] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0300] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0301] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0302] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0303] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A communication method applied to a base station, characterized in that, include: Send resource configuration information to the user equipment (UE), wherein the resource configuration information is used to indicate uplink physical resources; The system receives monitoring data sent by the UE based on the uplink physical resources. The monitoring data is used to monitor the Channel State Information (CSI) prediction module and / or CSI compression two-sided model in the base station and / or the UE.

2. The method according to claim 1, characterized in that, After receiving the monitoring data sent by the UE based on the uplink physical resources, the method further includes: Based on the monitoring data, the CSI prediction module and / or CSI compression two-sided model in the base station and / or the terminal device are monitored.

3. The method according to claim 1 or 2, characterized in that, Before receiving the monitoring data sent by the UE based on the uplink physical resources, the method further includes: Send a monitoring data reporting configuration to the UE, wherein the monitoring data reporting configuration is used to indicate the monitoring data to be reported and the reporting format of the monitoring data.

4. The method according to claim 3, characterized in that, The monitoring data reporting configuration includes, but is not limited to, at least one of the following parameters: reporting volume, compressed CSI auxiliary information, compressed CSI auxiliary information granularity, monitoring indicators, monitoring indicator value dimension, and / or monitoring indicator threshold.

5. The method according to claim 4, characterized in that, The indication methods for each parameter in the monitoring data reporting configuration include, but are not limited to, at least one of the following: direct indication, index indication, single bit indication, and / or bitmap indication.

6. The method according to claim 4 or 5, characterized in that, The reported data includes, but is not limited to, at least one of the following: uncompressed CSI, compressed CSI, UE-side monitoring index values, and / or UE-side monitoring results.

7. The method according to claim 6, characterized in that, When the reported quantity includes the UE-side monitoring results, the method further includes: Receive the CSI prediction module performance monitoring results sent by the UE.

8. The method according to claim 6, characterized in that, When the reported quantity includes the uncompressed CSI, the monitoring data includes, but is not limited to, at least one of the following: CSI-RS Resource Indicator (CRI), Rank Indicator (RI), Layer Indicator (LI), Precoding Matrix Indicator (PMI), and / or Channel Quality Indicator (CQI).

9. The method according to claim 8, characterized in that, When the reported quantity includes the compressed CSI, the monitoring data includes, but is not limited to, at least one of the following: CRI, RI, LI, compressed precoding matrix, and / or CQI.

10. The method according to claim 8, characterized in that, When the reported quantity includes the monitoring index value on the UE side, the monitoring data includes the monitoring index value calculated by the UE through predicted CSI and measured CSI, and the monitoring index corresponding to the monitoring index value is configured by the base station.

11. The method according to claim 8, characterized in that, When the reported quantity includes the monitoring results on the UE side, the monitoring data includes the performance judgment results of the CSI prediction module.

12. The method according to any one of claims 4 to 11, characterized in that, The compressed CSI auxiliary information includes, but is not limited to, at least one of the following: quantization error of the compressed precoding matrix or codebook adjustment value.

13. The method according to any one of claims 1 to 12, characterized in that, The resource configuration information includes, but is not limited to, at least one of the following methods: indicating the uplink physical resources corresponding to each monitoring data by configuring the reporting configuration for each monitoring data type; indicating the monitoring data type corresponding to the CSI reporting configuration associated with the CSI semi-continuous reporting trigger state and / or the CSI non-periodic reporting trigger state; and / or indicating the reporting order of monitoring data types by using downlink control information (DCI).

14. The method according to any one of claims 1 to 13, characterized in that, After sending resource configuration information to the UE, the method further includes: Receive quality information of each monitoring data sent by the UE.

15. The method according to claim 14, characterized in that, The quality information includes, but is not limited to, at least one of the following: the error between the precoded tensor obtained from PMI and the actual precoded tensor, candidate values ​​within a predefined range, and / or UE measurement values.

16. A communication method applied to the UE side, characterized in that, include: Determine uplink physical resources; Based on the uplink physical resources, monitoring data is reported to the base station. The monitoring data is used to monitor the CSI prediction module and / or CSI compression two-sided model in the base station and / or the terminal device.

17. The method according to claim 16, characterized in that, The determined uplink physical resources include: The system receives resource configuration information sent by the base station, which is used to indicate the uplink physical resources.

18. The method according to claim 16, characterized in that, The determined uplink physical resources include: The physical resources corresponding to each monitoring data are determined according to the reporting order of the predefined monitoring data types.

19. The method according to claim 16, characterized in that, The determined uplink physical resources include: The uplink physical resources are determined based on the CSI calculation time.

20. The method according to claim 19, characterized in that, The determination of the uplink physical resources based on CSI calculation time includes: Determine the delay time corresponding to each monitored data type based on the CSI calculation time; Based on the delay time corresponding to each monitoring data type, the reporting order of each monitoring data type is obtained; Based on the reporting order of the various monitoring data types, the uplink physical resources corresponding to each monitoring data type are determined.

21. The method according to claim 20, characterized in that, The process of determining the latency corresponding to each monitored data type based on the CSI calculation time includes: The sum of the specific symbol interval and the CSI calculation time corresponding to each monitoring data type is determined as the delay time corresponding to each monitoring data type. The specific symbol interval is the symbol interval between the last symbol of the DCI that triggers CSI reporting and the first or last symbol of the most recently measured CSI-RS resource.

22. The method according to claim 20 or 21, characterized in that, The smaller the delay time corresponding to each monitoring data type, the earlier the reporting order of each monitoring data type.

23. The method according to claim 19, characterized in that, The determination of the uplink physical resources based on CSI calculation time includes: The reporting order of each monitoring data type is determined based on the CSI calculation time corresponding to each monitoring data type. Based on the reporting order of the various monitoring data types, the uplink physical resources corresponding to each monitoring data type are determined.

24. The method according to claim 23, characterized in that, The shorter the CSI calculation time for each monitoring data type, the earlier the reporting order for that monitoring data type.

25. The method according to any one of claims 16 to 24, characterized in that, The method further includes: Historical channel information is obtained by measuring the CSI reference signal (CSI-RS). The monitoring data is calculated based on the historical channel information.

26. The method according to claim 25, characterized in that, When the monitoring data includes compressed CSI, the calculation of the monitoring data based on the historical channel information includes: Determine the computation time for the compressed CSI.

27. The method according to claim 26, characterized in that, The determination of the calculation time for the compressed CSI includes: The sum of the calculation time of the predicted CSI and a first calculation time offset is determined as the calculation time of the compressed CSI, wherein the first calculation time offset is determined by at least one of the following methods: based on UE capability, based on base station indication, and / or using a predefined value.

28. The method according to claim 26, characterized in that, The determination of the calculation time for the compressed CSI includes: The rounded value of the product of the predicted CSI calculation time and the offset factor is determined as the calculation time of the compressed CSI; wherein the offset factor is determined by at least one of the following methods: based on UE capability, based on base station indication, and / or using a predefined value.

29. The method according to claim 26, characterized in that, The determination of the calculation time for the compressed CSI includes: The computation time of the compressed CSI is determined based on at least one specific parameter, including but not limited to at least one of the following parameters: CSI prediction module computation time, encoder computation time, and / or quantizer computation time.

30. The method according to claim 29, characterized in that, The step of determining the calculation time of the compressed CSI based on at least one specific parameter includes: The sum of the calculation time of the CSI prediction module, the calculation time of the encoder, and the calculation time of the quantizer is taken as the calculation time of the compressed CSI.

31. The method according to claim 29, characterized in that, The step of determining the calculation time of the compressed CSI based on at least one specific parameter includes: The sum of the calculation time of the CSI prediction module and the calculation time of the encoder is used as the calculation time of the compressed CSI.

32. The method according to claim 29, characterized in that, The step of determining the calculation time of the compressed CSI based on at least one specific parameter includes: The sum of the quantizer calculation time and the encoder calculation time is used as the calculation time of the compressed CSI.

33. The method according to claim 29, characterized in that, The step of determining the calculation time of the compressed CSI based on at least one specific parameter includes: The encoder calculation time is used as the calculation time of the compressed CSI.

34. The method according to claim 26, characterized in that, The determination of the calculation time for the compressed CSI includes: The predefined value is used as the computation time of the compressed CSI, which is N times the CSI computation time under a specific codebook type, wherein N is included in the set {1, 2, 3, 4, 6}, and the specific codebook type includes, but is not limited to, at least one of the following types: Type II codebook, Type II port selection codebook, enhanced Type II codebook, enhanced Type II port selection codebook, and / or further enhanced Type II port selection codebook.

35. The method according to any one of claims 16 to 34, characterized in that, The method further includes: Determine the quality information of each monitoring data point; The quality information of each monitoring data is reported to the base station.

36. The method according to claim 35, characterized in that, The quality information includes, but is not limited to, at least one of the following: the error between the precoded tensor obtained from PMI and the actual precoded tensor, candidate values ​​within a predefined range, and / or UE measurement values.

37. The method according to claim 36, characterized in that, The quality information for measuring CSI can be obtained using, but is not limited to, at least one of the following UE measurements: CSI-RS received power (CSI-RSRP), CSI-RS received quality (CSI-RSRQ), CSI signal-to-interference-plus-noise ratio (CSI-SINR), received signal strength indicator (RSSI), and channel quality indicator (CQI).

38. The method according to any one of claims 16 to 37, characterized in that, The reporting of monitoring data to the base station based on the uplink physical resources includes: When the first preset condition is met, the monitoring data is reported according to the priority rules.

39. The method according to claim 38, characterized in that, When the monitoring data and non-monitoring data (CSI) share the priority rule, the relevant factors of the priority rule include, but are not limited to, at least one of the following: CSI reporting configuration ID, serving cell index, reporting content, reporting type, and / or the type of monitoring data carried.

40. The method according to claim 38, characterized in that, When the priority rule is applied to the monitoring data, the relevant factors of the priority rule include, but are not limited to, at least one of the following: CSI reporting configuration ID, serving cell index, monitored data type carried, and / or data volume index value, wherein each data volume index value corresponds to a monitoring data volume range.

41. The method according to claim 38, characterized in that, When the first preset condition is met, the monitoring data reported according to the priority rules includes: When multiple monitoring data are carried on a single Physical Uplink Shared Channel (PUSCH), the monitoring data includes a first part of data and a second part of data, wherein... The second part of each monitoring data is reported according to the priority rule, which is related to the reporting priority of the monitoring data and the data grouping within the second part of the data.

42. The method according to claim 41, characterized in that, The monitored data types are the first part of the data corresponding to the predicted CSI and the measured CSI, including but not limited to at least one of the following parameters: RI, complete CQI or the first CQI, and / or the number of interlayer non-zero amplitude coefficients. The corresponding second part of the data includes but is not limited to at least one of the following parameters: the second CQI and / or PMI.

43. The method according to claim 41, characterized in that, The monitored data type is the first part of the data corresponding to compressed CSI, including but not limited to at least one of the following parameters: RI, number of compressed precoding matrix elements, layer indicator of the bearer, and / or subspace indicator of the bearer. The corresponding second part of the data includes but is not limited to at least one of the following parameters: compressed precoding matrix, quantization error, and / or codebook adjustment value.

44. The method according to any one of claims 16 to 43, characterized in that, Before reporting monitoring data to the base station, the method further includes: The system receives monitoring data reporting configuration sent by the base station. The monitoring data reporting configuration is used to indicate the monitoring data to be reported and the reporting format of the monitoring data.

45. The method according to claim 44, characterized in that, The monitoring data reporting configuration includes, but is not limited to, at least one of the following parameters: reporting volume, compressed CSI auxiliary information, compressed CSI auxiliary information granularity, monitoring indicators, monitoring indicator value dimension, and / or monitoring indicator threshold.

46. ​​The method according to claim 45, characterized in that, The reported data includes, but is not limited to, at least one of the following: uncompressed CSI, compressed CSI, UE-side monitoring index values, and UE-side monitoring results.

47. The method according to claim 46, characterized in that, When the reported quantity includes the UE-side monitoring results, the method further includes: The performance monitoring results of the CSI prediction module are determined by comparing the monitoring index values ​​with the monitoring index thresholds indicated by the base station. The CSI prediction module performance monitoring results are sent to the base station.

48. A communication method applied to the UE side, characterized in that, include: Determine the downlink physical resources for reconstructing the CSI in order to receive the reconstructed CSI; The reconstructed CSI and monitoring data sent by the base station are associated.

49. The method according to claim 48, characterized in that, After determining the downlink physical resources for the reconstructed CSI, the method further includes: Based on the reconstructed CSI and the monitoring data, the CSI prediction module and / or CSI compression two-sided model in the base station and / or the terminal device are monitored.

50. The method according to claim 48 or 49, characterized in that, The determination of the downlink physical resources for the reconstructed CSI includes: Receive the downlink physical resource allocation configured by the base station.

51. The method according to claim 50, characterized in that, The associated reconstructed CSI and monitoring data sent by the base station include: The system receives association parameters sent by the base station, including but not limited to at least one of the following parameters: associated CSI-RS resource set index, associated CSI-RS resource index, associated compressed CSI reporting configuration ID, and / or compressed CSI reporting ID. The reconstructed CSI and the monitoring data are associated based on the association parameters.

52. The method according to claim 50 or 51, characterized in that, The method for determining the compressed CSI reporting ID includes, but is not limited to, at least one of the following methods: directly determined by the UE; the compressed CSI reporting ID triggered by the same DCI is determined by accumulating the ordinal number of the compressed CSI reports triggered by the same DCI based on the initial compressed CSI reporting ID; and / or directly indicated by the base station.

53. The method according to claim 51, characterized in that, The method for determining the initial compressed CSI reporting ID includes, but is not limited to, at least one of the following methods: determined by the UE, using a predefined value, and / or indicated by the base station.

54. The method according to claim 48 or 49, characterized in that, The determination of the downlink physical resources for the reconstructed CSI includes: The downlink physical resources are determined based on time-frequency correlation parameters, which are configured or predefined by the base station.

55. The method according to claim 54, characterized in that, The associated reconstructed CSI and monitoring data sent by the base station include: The reconstructed CSI and the monitoring data are associated based on the time-frequency correlation parameters.

56. The method according to claim 54 or 55, characterized in that, The time-frequency related parameters include, but are not limited to, at least one of the following parameters: the offset of the time slot where the physical resource is located relative to the time slot where the physical resource carrying the compressed CSI is located, the starting symbol of the physical resource in its time slot, the offset of the starting symbol of the physical resource relative to the starting symbol or the last symbol of the physical resource carrying the compressed CSI, the offset of the starting resource block of the physical resource relative to the starting resource block of the physical resource carrying the compressed CSI, the number of symbols occupied by the physical resource, and / or the number of resource blocks occupied by the physical resource.

57. The method according to any one of claims 48 to 56, characterized in that, The method further includes: Historical channel information is obtained by measuring CSI-RS, and compressed CSI is output based on the historical channel information; The compressed CSI is sent to the base station.

58. The method according to claim 57, characterized in that, Sending the compressed CSI to the base station includes: When the second preset condition is met, the compressed CSI is reported according to the priority rule.

59. The method according to claim 58, characterized in that, When the monitoring data and non-monitoring data (CSI) share the priority rule, the relevant factors of the priority rule include, but are not limited to, at least one of the following: CSI reports the configuration ID, serving cell index, report content, report type, and / or the data type of the monitoring data it carries.

60. The method according to claim 58, characterized in that, When the priority rule is applied to the monitoring data, the relevant factors of the priority rule include, but are not limited to, at least one of the following: CSI reporting configuration ID, serving cell index, monitored data type carried, and / or data volume index value, wherein each data volume index value corresponds to a monitoring data volume range.

61. The method according to claim 58, characterized in that, When the second preset condition is met, the compressed CSI is reported according to the priority rules, including: When monitoring data including the compressed CSI is carried on a PUSCH, the compressed CSI includes a first part of data and a second part of data, wherein, The second part of the data is reported according to the priority rules, which are related to the compressed CSI reporting priority and the data grouping within the second part of the data.

62. The method according to claim 59, characterized in that, The first part of the data includes, but is not limited to, at least one of the following parameters: RI, number of compressed precoding matrix elements, layer indicator of the bearer, and / or subspace indicator of the bearer. The second part of the data includes, but is not limited to, at least one of the following parameters: compressed precoding matrix, quantization error, and / or codebook adjustment value.

63. The method according to any one of claims 48 to 62, characterized in that, The method further includes: Receive the quality information of the reconstructed CSI sent by the base station.

64. The method according to claim 63, characterized in that, The quality information includes, but is not limited to, at least one of the following: the error between the precoded tensor obtained by PMI and the actual precoded tensor, and / or candidate values ​​within a predefined range.

65. A communication method applied to a base station, characterized in that, include: Indicate the configured downlink physical resources to the UE, and send a reconfiguration CSI based on the downlink physical resources.

66. The method according to claim 65, characterized in that, The step of instructing the UE to configure downlink physical resources and sending a reconfiguration CSI based on the downlink physical resources includes: Send downlink physical resource allocation configuration for the reconstructed CSI to the UE; The reconstructed CSI and associated parameters are sent to the UE, wherein the associated parameters are used to associate the reconstructed CSI and monitoring data.

67. The method according to claim 65, characterized in that, The step of instructing the UE to configure downlink physical resources and sending a reconfiguration CSI based on the downlink physical resources includes: Time-frequency related parameters are sent to the UE, and the time-frequency related parameters are used to determine the downlink physical resources; The reconstructed CSI is sent based on the downlink physical resources.

68. The method according to claim 66 or 67, characterized in that, Before sending the reconstructed CSI and associated parameters to the UE, the method further includes: Receive the compressed CSI sent by the UE; The compressed CSI is used as input to the base station-side model, and the reconstructed CSI is output.

69. The method according to any one of claims 66 to 68, characterized in that, The method further includes: The quality information of the reconstructed CSI is sent to the UE.

70. A wireless communication device, comprising: A processor and a memory, the memory being used to store a computer program, and the processor being used to invoke and run the computer program stored in the memory to perform the method as described in any one of claims 1 to 69.

71. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes instructions that, when executed, cause the method according to any one of claims 1 to 69 to be implemented.

Images