METHOD FOR PROCESSING CHANNEL STATE INFORMATION, TERMINAL, BASE STATION, AND COMPUTER PROGRAM

Abstract

The present application provides a channel state information processing method, a terminal, a base station, and a medium, and the channel state information processing method includes: packet pre-processing initial channel state information in a target dimension to obtain multiple first channel state information; encoding the multiple first channel state information to generate second channel state information; and feeding back the second channel state information.

Description

[Technical field]

[0001] (CROSS REFERENCE TO RELATED APPLICATIONS) This application was filed on March 22, 2022 and has application number 202210286749.1 The present application claims priority from Chinese patent application Ser. No. 09 / 239,363, the entire disclosure of which is incorporated herein by reference.

[0002] The present application relates to the field of communication technology, in particular to a method for processing channel state information, a terminal, a base station, and a computer program Regarding. [Background technology]

[0003] From Long Term Evolution (LTE) to New Radio (NR) access technology, multi-antenna technology has always been one of the key technologies in communication standards. With the improvement of communication standard indicators, how to obtain channel state information (CSI) more accurately has become the key to improving the performance of multi-antenna technology. Summary of the Invention

[0004] In a first aspect, an embodiment of the present disclosure provides a method for processing channel state information, the method including: packet pre-processing initial channel state information in a target dimension to obtain a plurality of first channel state information, encoding the plurality of first channel state information to generate second channel state information, and feeding back the second channel state information.

[0005] In a second aspect, an embodiment of the present disclosure provides a method for processing channel state information, the method including: receiving second channel state information, decoding the second channel state information to obtain a plurality of third channel state information, the plurality of third channel state information having one-to-one correspondence with a plurality of first channel state information obtained by packet pre-processing initial channel state information in a target dimension, and generating target channel state information according to the plurality of third channel state information.

[0006] In a third aspect, an embodiment of the present disclosure provides a terminal, the terminal including: one or more processors; a memory in which one or more programs are stored, the one or more processors implementing the channel state information processing method according to the first aspect of the embodiment of the present disclosure when the one or more programs are executed by the one or more processors; One or more A processor connected between the processor and the memory, One or more and one or more I / O interfaces configured to provide information interactivity between the processor and the memory.

[0007] In a fourth aspect, an embodiment of the present disclosure provides a base station, the base station including: one or more processors; a memory storing one or more programs, the one or more programs being executed by the one or more processors to cause the one or more processors to realize the channel state information processing method according to the second aspect of the embodiment of the present disclosure; One or more A processor connected between the processor and the memory, One or more and one or more I / O interfaces configured to provide information interactivity between the processor and the memory.

[0008] In a fifth aspect, an embodiment of the present disclosure provides a computer-readable medium having a computer program stored thereon, the computer-readable medium being adapted to implement any one of the above methods for processing channel state information when executed by a processor.

[0009] The embodiments of the present disclosure provide a terminal and a channel state information processing method applied to the terminal, a base station and a channel state information processing method applied to the base station, and a computer-readable medium. In the embodiments of the present disclosure, the terminal performs packet pre-processing of initial channel state information in a target dimension to obtain a plurality of first channel state information; Multiple Feedback second channel state information obtained by encoding the first channel state information. The base station receives and decodes the second channel state information to obtain target channel state information as an estimation of the initial channel state information, thereby completing the feedback of the channel state information. Here, the size of the channel state information processed by the encoder and the decoder is smaller than the size of the initial channel state information, so that the amount of input data of the encoder and the decoder can be reduced, which is favorable for simplifying the model of the encoder and the decoder, and under the premise of ensuring the accuracy of the channel state information feedback, the overhead of the terminal calculation and the channel state information feedback can be reduced.

[0010] In order to more clearly describe the technical solutions in the present disclosure, the drawings used in some embodiments of the present disclosure will be briefly described below. It is clear that the drawings in the following description are only drawings of some embodiments of the present disclosure. Those skilled in the art can obtain other drawings based on these drawings. In addition, the drawings in the following description can be regarded as schematic diagrams, and do not limit the actual dimensions of the products, the actual flow of the methods, the actual timing of the signals, etc., according to the embodiments of the present disclosure. [Brief description of the drawings]

[0011] [Figure 1] 4 is a flow chart of a method for processing channel state information according to some embodiments. [Diagram 2] 11 is a flowchart of another channel state information processing method according to some embodiments. [Diagram 3] 10 is a flowchart of yet another method for processing channel state information according to some embodiments. [Figure 4] 10 is a flowchart of yet another method for processing channel state information according to some embodiments. [Diagram 5] 10 is a flowchart of yet another method for processing channel state information according to some embodiments. [Figure 6] 11 is a flowchart of another channel state information processing method according to some embodiments. [Figure 7] FIG. 2 is a block diagram of a terminal according to some embodiments. [Figure 8] FIG. 2 is a block diagram of a base station according to some embodiments. [Figure 9] FIG. 1 is a block diagram of a computer-readable medium according to some embodiments. [Figure 10] FIG. 2 is a schematic diagram of packet pre-processing according to some embodiments. [Figure 11] FIG. 13 is a schematic diagram of another packet pre-processing method according to some embodiments. [Figure 12] FIG. 13 is a schematic diagram of yet another packet pre-processing method according to some embodiments. [Figure 13] FIG. 13 is a schematic diagram of yet another packet pre-processing method according to some embodiments. [Figure 14] FIG. 13 is a schematic diagram of yet another packet pre-processing method according to some embodiments. [Figure 15] FIG. 13 is a schematic diagram of yet another packet pre-processing method according to some embodiments. [Figure 16] FIG. 13 is a schematic diagram of yet another packet pre-processing method according to some embodiments. [Figure 17A] FIG. 13 is a schematic diagram of yet another packet pre-processing method according to some embodiments. [Figure 17B] FIG. 13 is a schematic diagram of yet another packet pre-processing method according to some embodiments. [Figure 18]FIG. 13 is a schematic diagram of yet another packet pre-processing method according to some embodiments. [Figure 19] FIG. 13 is a schematic diagram of yet another packet pre-processing method according to some embodiments. [Figure 20] FIG. 13 is a schematic diagram of yet another packet pre-processing method according to some embodiments. [Figure 21] FIG. 13 is a schematic diagram of yet another packet pre-processing method according to some embodiments. [Figure 22] FIG. 13 is a schematic diagram of yet another packet pre-processing method according to some embodiments. [Figure 23] FIG. 13 is a schematic diagram of yet another packet pre-processing method according to some embodiments. [Figure 24] 1 is a schematic diagram of a channel state information feedback system according to some embodiments; [Diagram 25] 1 is a schematic diagram of feedback of polarization phase aiding information according to some embodiments; DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0012] In order for those skilled in the art to better understand the technical solutions of the embodiments of the present disclosure, the channel state information processing method, terminal, base station, and computer-readable medium provided by the embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings.

[0013] Exemplary embodiments will now be described in more detail with reference to the accompanying drawings, in which: FIG. 1 is a schematic diagram of a method for manufacturing a semiconductor device according to an embodiment of the present invention;

[0014] Hereinafter, the terms "first" and "second" are merely for descriptive purposes and should not be understood as indicating or implying the relative importance or quantity of the indicated technical features. Where not inconsistent, each embodiment and each feature in the embodiments of the present disclosure may be combined with each other.

[0015] As used herein, the term "and / or" includes any and all combinations of one or more of the associated listed items.

[0016] The terms used herein are used only to describe certain embodiments and are not intended to limit the present disclosure. As used herein, the singular forms "a," "an," and "the" are intended to include the plural, unless the context indicates otherwise. It is also understood that the terms "comprising" and / or "comprising" used herein specify the presence of a stated feature, whole, step, operation, element, and / or assembly, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, assemblies, and / or groups thereof.

[0017] Unless otherwise limited, all terms used herein (including technical and scientific terms) have the same meaning as commonly understood by those skilled in the art. It is also understood that terms limited to common dictionaries should be interpreted as having a meaning consistent with the meaning in the context of the relevant art and the embodiments of the present disclosure, and are not interpreted as having an ideal or excessive form meaning, unless expressly limited herein.

[0018] In a multi-antenna system, the downlink channel state information fed back from the user equipment (UE) to the base station (BS) is affected by the large antenna array, and the overhead of feeding back the downlink channel state information is positively correlated with the number of antennas. The base station may include a gNB, eNB, or Node B. In a complex environment scenario, the base station may receive feedback from the UE. Depending on the backed-up codebook information, the accuracy of recovering the downlink channel state information also needs to be improved.

[0019] Artificial intelligence (AI) technology has demonstrated powerful feature extraction and learning capabilities in the fields of computer vision, natural language processing, and voice signal processing. In some related technologies, the introduction of artificial intelligence technology into wireless communication systems breaks through the bottlenecks of some related technologies in wireless communication systems. However, in AI-based CSI feedback technology, the amount of data input to the neural network is relatively large, the model is relatively complex, and the number of parameters and the amount of floating-point calculations of the neural network are greatly increased, which increases the overhead cost of terminal processing and CSI feedback.

[0020] Some embodiments of the present disclosure provide a lightweight AI concept to preprocess complex input data of the terminal encoder, thereby reducing the overhead costs of terminal computation and CSI feedback, under the premise of ensuring the accuracy of the CSI feedback.

[0021] In a first aspect, referring to FIG. 1, some embodiments of the present disclosure provide a method for processing channel state information, the method including steps S11 to S13.

[0022] S11, packet pre-processing initial channel state information in a target dimension to obtain a plurality of first channel state information.

[0023] In some embodiments, the terminal obtains initial channel state information according to received channel state reference information, such as a channel state information reference signal (CSI-RS). In the embodiments of the present disclosure, the initial channel state information may be time domain channel information or frequency domain channel information, and the embodiments of the present disclosure are not particularly limited thereto.

[0024] The embodiment of the present disclosure does not particularly limit the terminal, and the terminal may be any of a mobile phone, a data card, a notebook computer, and various manufacturing equipment in a factory.

[0025] In some embodiments of the present disclosure, the initial channel state information has multiple dimensions. For example, the multiple dimensions include a frequency domain dimension, a spatial domain dimension, a layer dimension, and a time domain dimension. The embodiments of the present disclosure do not particularly limit the target dimension. In some embodiments, the target dimension includes at least one of a frequency domain dimension, a spatial domain dimension, a layer dimension, and a time domain dimension.

[0026] For example, the target dimension may be any one of a frequency domain dimension, a spatial domain dimension, a layer dimension, and a time domain dimension. Correspondingly, packet pre-processing the initial channel state information in the target dimension refers to packet pre-processing the initial channel state information in one dimension of a frequency domain dimension, a spatial domain dimension, a layer dimension, and a time domain dimension. Also, for example, the target dimension may be a combination of at least two of a frequency domain dimension, a spatial domain dimension, a layer dimension, and a time domain dimension. Correspondingly, packet pre-processing the initial channel state information in the target dimension refers to packet pre-processing the initial channel state information in at least two dimensions of a frequency domain dimension, a spatial domain dimension, a layer dimension, and a time domain dimension.

[0027] The embodiments of the present disclosure do not particularly limit how the initial channel state information is packet pre-processed in the target dimension.

[0028] In an embodiment of the present disclosure, after packet pre-processing of the initial channel state information in the target dimension, the dimension of the obtained first channel state information in the target dimension is smaller than the dimension of the initial channel state information in the target dimension.

[0029]

number

[0030] S12, encoding the plurality of first channel state information to generate second channel state information. In the embodiment of the present disclosure, in S12, encoding the plurality of first channel state information is inputting the first channel state information to an encoder configured by a neural network, and encoding and compressing the first channel state information. Compared with the case where the initial channel state information is input to the encoder, when the first channel state information is input to the encoder, the amount of input data to the encoder can be reduced.

[0031] S13, feeding back the second channel state information. In the channel state information processing method provided by the embodiments of the present disclosure, after the initial channel state information is packet pre-processed in a target dimension, a plurality of first channel state information are obtained, and the dimension of the first channel state information is smaller than that of the initial channel state information, so that the amount of input data to the encoder can be reduced, and the encoder model can be simplified, which is advantageous for reducing the overhead of terminal calculation and feedback of channel state information under the premise of ensuring the accuracy of channel state information feedback.

[0032] In some embodiments, referring to FIG. 2, S11 includes S111. S111, divide a plurality of resource granularities in a target dimension of initial channel state information into a plurality of resource granularity groups to obtain a plurality of first channel state information, where the plurality of resource granularity groups correspond one-to-one to the plurality of first channel state information.

[0033] In an embodiment of the present disclosure, the initial channel state information has multiple resource granularities in a target dimension, and the resource granularity is the smallest unit by which the initial channel state information is divided in the target dimension. In an embodiment of the present disclosure, when the target dimensions are different, the resource granularities are also different. For example, when the target dimension is a frequency domain dimension, the resource granularity is a frequency domain granularity; when the target dimension is a spatial domain dimension, the resource granularity is an antenna port or a polarization direction; when the target dimension is a layer dimension, the resource granularity is a layer granularity; and when the target dimension is a time domain dimension, the resource granularity is a time domain granularity. Correspondingly, when the target dimension is a frequency domain dimension, the initial channel state information is packet pre-processed in the frequency domain dimension to obtain a plurality of frequency domain granularity groups composed of frequency domain granularity; when the target dimension is a spatial domain dimension, the initial channel state information is packet pre-processed in the spatial domain dimension to obtain a plurality of antenna port groups composed of antenna ports or a plurality of polarization direction groups composed of polarization directions; when the target dimension is a layer dimension, the initial channel state information is packet pre-processed in the layer dimension to obtain a plurality of layer granularity groups composed of layer granularity; when the target dimension is a time domain dimension, the initial channel state information is packet pre-processed in the time domain dimension to obtain a plurality of time domain granularity groups composed of time domain granularity.

[0034]

number

[0035] The embodiment of the present disclosure does not particularly limit how to divide the multiple resource granularities in the target dimension of the initial channel state information into multiple resource granularity groups. It should be noted that the packet scheme provided by the embodiment of the present disclosure is applicable to the target dimension such as frequency domain dimension, spatial domain dimension, layer dimension, or time domain dimension.

[0036] In some embodiments, in S111, dividing a plurality of resource granularities in a target dimension of the initial channel state information into a plurality of resource granularity groups to obtain a plurality of first channel state information includes:

[0037] According to the packet granularity number, among the multiple resource granularities in the target dimension of the initial channel state information, the resource granularities of each consecutive packet granularity number are divided into one resource granularity group to obtain multiple first channel state information, and the packet granularity number characterizes the number of resource granularities in the resource granularity group.

[0038] It should be noted that in the embodiment of the present disclosure, the indices of multiple resource granularities in the target dimension of the initial channel state information are numbered sequentially, and consecutive resource granularities are index consecutive resource granularities.

[0039]

number

[0040] In some other embodiments, in S111, dividing a plurality of resource granularities in a target dimension of the initial channel state information into a plurality of resource granularity groups to obtain a plurality of first channel state information includes:

[0041] According to the packet granularity number, among the multiple resource granularities in the target dimension of the initial channel state information, the resource granularities of the packet granularity number arranged at equal intervals are divided into one resource granularity group to obtain multiple first channel state information, where the packet granularity number characterizes the number of resource granularities in the resource granularity group.

[0042] In addition, in an embodiment of the present disclosure, the indices of multiple resource granularities in the target dimension of the initial channel state information are numbered sequentially, and the multiple resource granularities arranged at equal intervals are multiple resource granularities whose indices are equally spaced.

[0043]

number

[0044] In the embodiment of the present disclosure, dividing the resource granularity of each equally spaced packet granularity number into one resource granularity group includes Adjacent The span of resource granularity can be increased, which can improve the processing performance of channel state information.

[0045] In some other embodiments, in S111, dividing a plurality of resource granularities in a target dimension of the initial channel state information into a plurality of resource granularity groups to obtain a plurality of first channel state information includes: dividing a plurality of resource blocks in the initial channel state information into a plurality of resource granularity groups according to a packet granularity number to obtain a plurality of first channel state information; the packet granularity number characterizes the number of resource granularities in the resource granularity group; the resource block includes a portion of the plurality of resource granularities in the target dimension of the initial channel state information, and the sum of the number of all resource blocks divided into one resource granularity group is equal to the packet granularity number.

[0046] It should be noted that in the embodiment of the present disclosure, the indices of the multiple resource granularities in the target dimension of the initial channel state information are numbered sequentially, and a resource block includes multiple resource granularities with consecutive indexes.

[0047] The embodiment of the present disclosure does not particularly limit the number of resource granularities in each resource block. The number of resource blocks divided into one resource granularity group may be two or more. For example, when each resource granularity group includes six resource granularities, if each resource block includes three resource granularities, the number of resource blocks divided into one resource granularity group is two, and if each resource block includes two resource granularities, the number of resource blocks divided into one resource granularity group is three. The embodiment of the present disclosure does not particularly limit this.

[0048] The embodiments of the present disclosure do not particularly limit how resource blocks are divided into resource granularity groups.

[0049] In some examples, dividing the plurality of resource blocks in the initial channel state information into a plurality of resource granularity groups according to the packet granularity number and obtaining a plurality of pieces of first channel state information includes dividing the resource granularity of each of a plurality of equally spaced resource blocks among the plurality of resource blocks in the initial channel state information into one resource granularity group according to the packet granularity number and obtaining a plurality of pieces of first channel state information.

[0050] It should be noted that in the embodiment of the present disclosure, the indexes of the multiple resource granularities in the target dimension of the initial channel state information are numbered in sequence, and each of several consecutive resource granularities is divided into one resource block. The indexes of the multiple resource blocks are also numbered in sequence, and the resource blocks arranged at equal intervals refer to multiple resource blocks whose indexes are at equal intervals.

[0051] In the embodiment of the present disclosure, dividing each equally-spaced resource block into one resource granularity group includes: Adjacent The span of resource granularity can be increased, which can improve the processing performance of channel state information.

[0052] In some other examples, dividing the multiple resource blocks in the initial channel state information into multiple resource granularity groups according to the packet granularity number and obtaining multiple pieces of first channel state information includes dividing the resource granularity in the multiple resource blocks into multiple resource granularity groups in a head and tail combination manner according to the packet granularity number and obtaining multiple pieces of first channel state information.

[0053] It should be noted that in the embodiment of the present disclosure, the indexes of the multiple resource granularities in the target dimension of the initial channel state information are numbered in sequence, and each successive several resource granularities are divided into one resource block. The multiple resource blocks are arranged in a queue in sequence. Dividing the resource granularities in the multiple resource blocks into multiple resource granularity groups in a head-to-tail combination manner refers to dividing the resource granularity in the first resource block from the head of the queue and the resource granularity in the first resource block from the tail into one resource granularity group, dividing the resource granularity in the second resource block from the head of the queue and the resource granularity in the second resource block from the tail into one resource granularity group, etc. to obtain multiple resource granularity groups, and thus obtain multiple first channel state information.

[0054] In the embodiment of the present disclosure, dividing a plurality of resource blocks into one resource granularity group in the manner of a head and tail combination is Adjacent The span of resource granularity can be increased, which can improve the processing performance of channel state information.

[0055] In some other embodiments, in S111, dividing a plurality of resource granularities in a target dimension of the initial channel state information into a plurality of resource granularity groups to obtain a plurality of first channel state information includes:

[0056] According to the packet granularity number, among the multiple resource granularities in the target dimension of the initial channel state information, the resource granularities of each packet granularity number are randomly divided into one resource granularity group to obtain multiple pieces of first channel state information. The packet granularity number characterizes the number of resource granularities in the resource granularity group.

[0057] The embodiment of the present disclosure does not particularly limit how the resource granularities of each packet granularity number are randomly divided into one resource granularity group. For example, multiple resource granularities may be randomly ordered and then divided into multiple resource granularity groups in the manner of consecutive packets, or multiple resource granularities may be randomly obtained to form one resource granularity group.

[0058] In the embodiments of the present disclosure, randomly dividing the resource granularity of each packet granularity number into one resource granularity group can improve the processing performance of the channel state information. In some examples, among the multiple resource granularities in the target dimension of the initial channel state information according to the packet granularity number, randomly dividing the resource granularity of each packet granularity number into one resource granularity group, and obtaining the multiple first channel state information includes randomly ordering the multiple resource granularities, obtaining a packet waiting sequence, and dividing the resource granularity of each successive packet granularity number into one resource granularity group according to the packet waiting sequence.

[0059] In some scenarios, the number of resource granularities in the target dimension of the initial channel state information is not divisible by the number of resource granularities to be divided into resource granularity groups, in which case the number of resource granularities in each resource granularity group may be made the same by padding the resource granularities.

[0060] Based on this, in some embodiments, in S111, dividing a plurality of resource granularities in a target dimension of the initial channel state information into a plurality of resource granularity groups to obtain a plurality of first channel state information further includes filling resource granularities when the number of resource granularities in the target dimension of the initial channel state information is not divisible by the number of packet granularities, where the number of packet granularities characterizes the number of resource granularities in the resource granularity group.

[0061] It should be noted that after the resource granularity is filled, the total number of resource granularities is divisible by the number of resource granularities to be divided into resource granularity groups. The embodiment of the present disclosure does not particularly limit how the resource granularity is filled.

[0062] By filling the resource granularity, it is possible to ensure that the number of resource granularity in each divided resource granularity group is the same, and it is possible to avoid the size of the data input to the encoder becoming irregular, and thus to avoid the size of the data input to the encoder being too small, which affects the processing performance of the channel state information, due to the remainder being too small when dividing the number of resource granularity in the target dimension of the initial channel state information by the number of resource granularity to be divided into resource granularity groups.

[0063] In some embodiments, filling the resource granularity includes filling a set number of resource granularities in the first target resource granularity group according to a number of waiting to be filled, where the number of waiting to be filled is a number of resource granularities that need to be filled.

[0064] The first target resource granularity group is a resource granularity group that contains less resource granularity numbers than packet granularity numbers before filling the resource granularities.

[0065] In some examples, the multiple resource granularity groups are numbered in sequence, and the first target resource granularity group is the resource granularity group with the highest number (i.e., penultimate). In some examples, the multiple resource granularities are packetized, the multiple resource granularity groups are obtained in sequence, and the first target resource granularity group is the last resource granularity group obtained. The embodiments of the present disclosure are not particularly limited thereto.

[0066] In the embodiment of the present disclosure, the setting resource granularity is an arbitrary value for filling. For example, the setting resource granularity is 0. The embodiment of the present disclosure does not particularly limit the setting resource granularity.

[0067]

number

[0068] In some other embodiments, filling the resource granularity includes duplicating the first target resource granularity by the number of waiting to be filled according to the number of waiting to be filled, and filling the first target resource granularity group, where the first target resource granularity is one of a plurality of resource granularities in a target dimension of the initial channel state information, and the number of waiting to be filled is the number of resource granularities that need to be filled.

[0069] In an embodiment of the present disclosure, the first target resource granularity is any one of a plurality of resource granularities in a target dimension of the initial channel state information. For example, the plurality of resource granularities are numbered in sequence, and the first target resource granularity may be the first resource granularity, the second resource granularity, or the last resource granularity. The embodiment of the present disclosure is not particularly limited thereto.

[0070]

number

[0071] In some other embodiments, filling the resource granularities includes duplicating the resource granularities in the sequence waiting to be filled to fill the first target resource granularity group according to a number of waiting to be filled, where the sequence includes the number of resource granularities waiting to be filled, and the number of waiting to be filled is the number of resource granularities that need to be filled.

[0072] The embodiment of the present disclosure is not particularly limited with respect to the refilling waiting sequence. For example, a refilling waiting sequence may be configured by randomly selecting a plurality of resource granularities, or a refilling waiting sequence may be configured by a plurality of resource granularities that are consecutive in forward order starting from a specific resource granularity, or a refilling waiting sequence may be configured by a plurality of resource granularities that are consecutive in reverse order starting from a specific resource granularity.

[0073] In some examples, replicating the resource granularities in the sequence waiting to be filled and filling it into a first target resource granularity group according to the number of waiting to be filled includes configuring the resource granularities of the number of waiting to be filled in successive forward or reverse order starting from a second target resource granularity as a sequence waiting to be filled, and replicating the resource granularities in the sequence waiting to be filled and filling it into the first target resource granularity group.

[0074] In some examples, the multiple resource granularities are numbered sequentially, and the second target resource granularity is one of the first resource granularity, the penultimate resource granularity (i.e., the last resource granularity), and the penultimate resource granularity in the multiple resource granularities in the target dimension of the initial channel state information. The embodiments of the present disclosure do not limit the second target resource granularity.

[0075]

number

[0076]

number

[0077]

number

[0078] In some embodiments, filling the resource granularity includes filling the second target resource granularity group in duplicate according to a number of pending fills, the number being the number of resource granularities that need to be filled.

[0079] In the embodiment of the present disclosure, duplicating and filling the second target resource granularity group refers to duplicating some resource granularities in a resource granularity group preceding the second target resource granularity group as resources in the second target resource granularity group.

[0080] In the embodiment of the present disclosure, a plurality of resource granularities in the target dimension of the initial channel state information may be packetized to obtain at least one resource granularity group, and then the second target resource granularity group may be filled in a redundant manner, so that the number of resource granularities in each resource granularity group finally obtained is the same. Alternatively, a plurality of resource granularities in the target dimension of the initial channel state information may be packetized, and at the same time, the second target resource granularity group may be filled in a redundant manner to obtain a plurality of resource granularity groups. The embodiment of the present disclosure is not particularly limited thereto.

[0081] In an embodiment of the present disclosure, one or more second target resource granularity groups may be filled in an overlapping manner. The embodiment of the present disclosure is not particularly limited thereto. In an embodiment of the present disclosure, a plurality of consecutive resource granularity groups may be filled in an overlapping manner, and a plurality of incomplete consecutive resource granularity groups may also be filled in an overlapping manner. The embodiment of the present disclosure is not particularly limited thereto.

[0082] The embodiment of the present disclosure does not particularly limit how the second target resource granularity group is filled with overlaps.

[0083] In some embodiments, filling the resource granularity includes determining an overlapping filling position and an overlapping resource dimension of at least one second target resource granularity group according to the number of waiting to be filled, and overlappingly filling each second target resource granularity group according to the overlapping filling position and overlapping resource dimension of each second target resource granularity group.

[0084] The overlap filling position characterizes the position of the second target resource granularity group, and the overlap resource dimension characterizes the number of resource granularities that need to be filled in the second target resource granularity group.

[0085] In some examples, determining the overlapping filling positions and overlapping resource dimensions of at least one second target resource granularity group according to the number of waiting to be filled includes determining a first number of consecutive second target resource granularity groups according to the number of waiting to be filled, each of the second target resource granularity groups having an overlapping resource dimension that is a second number, and the product of the first number and the second number is equal to the number of waiting to be filled.

[0086] According to the overlapping filling position and overlapping resource dimension of each second target resource granularity group, overlapping and filling each second target resource granularity group includes respectively duplicating the last second number of resource granularities in a resource granularity group preceding each second target resource granularity group, and making it the first second number of resource granularities in each second target resource granularity group.

[0087] In an embodiment of the present disclosure, the first number of consecutive second target resource granularity groups may be any first number of consecutive resource granularity groups among the multiple resource granularity groups, although the embodiment of the present disclosure is not particularly limited thereto.

[0088] In an embodiment of the present disclosure, the second number may be any of 1, 2, ..., a number of waiting to be filled. The embodiment of the present disclosure is not particularly limited thereto.

[0089] In some examples, the first number is equal to the number of waiting to be filled, and the second number is 1. Duplicating the second number of resource granularities from the end in one resource granularity group before each second target resource granularity group as the second number of resource granularities from the beginning in each second target resource granularity group includes duplicating the last resource granularity in one resource granularity group before each second target resource granularity group as the first resource granularity in each second target resource granularity group.

[0090]

number

[0091] Duplicating and filling each second target resource granularity group according to the overlapping filling position and overlapping resource dimension of each second target resource granularity group includes duplicating the resource granularity of the last number of waiting to be filled in one resource granularity group preceding the second target resource granularity group according to the number of waiting to be filled, and making it the resource granularity of the first number of waiting to be filled in the second target resource granularity group.

[0092]

number

[0093] In some other examples, determining the overlapping filling position and the overlapping resource dimension of at least one second target resource granularity group according to the number of waiting for filling includes determining a plurality of second target resource granularity groups and an overlapping resource dimension of each second target resource granularity group, and the overlapping resource dimensions of the at least two second target resource granularity groups are mutually different. different.

[0094] In the embodiment of the present disclosure, when the overlapping resource dimensions of at least two second target resource granularity groups are different from each other, the multiple second target resource granularities may be multiple consecutive resource granularity groups or multiple incompletely consecutive resource granularity groups. The embodiment of the present disclosure does not particularly limit this. The overlapping resource dimensions of each second target resource granularity group may be any value as long as the sum of the overlapping resource dimensions of all the second target resource granularity groups is equal to the number of waiting for filling. The embodiment of the present disclosure also does not particularly limit this.

[0095]

number

[0096] In some embodiments, referring to FIG. 3, before S11, the processing method further includes S14.

[0097] S14, receiving packet signaling, where the packet signaling includes packet parameters.

[0098] On this basis, with reference to FIG. 3, S11 includes S112. S112, packet pre-process the initial channel state information according to the packet parameters in a target dimension to obtain a plurality of first channel state information.

[0099] In some embodiments, the packet parameters include: A packet identification bit is used to identify whether the initial channel state information needs to be packetized and processed; a divisibility identification bit used to identify whether the number of resource granularities in the target dimension of the initial channel state information is divisible by the number of packet granularities; Packet type information is used to indicate a packet type for packet pre-processing the initial channel state information; a packet granularity number used to indicate the number of resource granularities in a resource granularity group obtained by packetizing the initial channel state information; When overlapping and filling, the overlapping filling position and overlapping resource of the second target resource granularity group and resource granularity filling information used to indicate dimensions.

[0100] In some embodiments, referring to FIG. 4, the processing method further includes S151 and S152.

[0101] S151, configuring a first packet arrangement list, where the first packet arrangement list includes at least one packet scheme for packet pre-processing the initial channel state information.

[0102] S152, information on the first packet arrangement list is fed back. In some embodiments, referring to FIG. 5, the processing method further includes S161 and S162.

[0103] S161, antenna polarization phase auxiliary information is obtained. S162, the polarization phase auxiliary information is fed back.

[0104] In some embodiments, the processing method further includes the step of transmitting a packet index, the packet index characterizing a correspondence relationship between the plurality of first channel state information and the plurality of resource granularities in the initial channel state information.

[0105] In some embodiments, when the initial channel state information is serialised and packet pre-processed in sequence, and then the first channel state information is coded and fed back in sequence, or when the initial channel state information is packet pre-processed in parallel, and then the first channel state information is coded and fed back in parallel, a packet index is sent to the base station, so that the base station recovers the initial channel state information according to the packet index.

[0106] In some embodiments, when the initial channel state information is serialised and packetised in sequence and then the first channel state information is coded and fed back in sequence, a packet index is not sent to the base station; when the initial channel state information is packetised in parallel and then the first channel state information is coded and fed back in parallel, a packet index is sent to the base station.

[0107] In some embodiments, when the initial channel state information is serialised and packet pre-processed in sequence, and then the first channel state information is coded and fed back in sequence, the packet index may be selected not to be sent to the base station, or the packet index may be sent to the base station. When the initial channel state information is packet pre-processed in parallel, and then the first channel state information is coded and fed back in parallel, the packet index is sent to the base station.

[0108] In a second aspect, referring to Fig. 6, an embodiment of the present disclosure provides another channel state information processing method, which includes S21-S23.

[0109] S21, receiving second channel state information. S22, decode the second channel state information to obtain a plurality of third channel state information, and the plurality of third channel state information corresponds one-to-one to the plurality of first channel state information obtained by packet pre-processing the initial channel state information in a target dimension.

[0110] S23, generating target channel state information according to the plurality of third channel state information. In an embodiment of the present disclosure, the base station can receive and decode the second channel state information to obtain target channel state information, which is an estimation of the initial channel state information, thereby completing the feedback of the channel state information.

[0111] The embodiment of the present disclosure does not particularly limit the base station, For example, the base station may be any of a macro base station, a micro base station, a home base station, and a pico base station.

[0112] In S23, the base station may generate target channel state information from the plurality of third channel state information according to a packet manner in which the terminal packet pre-processes the initial channel state information in a target dimension, where the dimension of the obtained target channel state information is the same as the dimension of the initial channel state information, and the target channel state information is an approximation or estimation of the initial channel state information.

[0113] Since the size of the channel state information processed by the encoder and the decoder is smaller than the size of the initial channel state information, the input data amount of the encoder and the decoder can be reduced, which is favorable for simplifying the models of the encoder and the decoder, and realizes reducing the overhead of terminal calculation and channel state information feedback under the premise of ensuring the accuracy of the channel state information feedback.

[0114] In some embodiments, in S23, generating target channel state information in response to the plurality of third channel state information includes obtaining the target channel state information in response to the plurality of third channel state information and a packet scheme for packet pre-processing the initial channel state information in a target dimension.

[0115] The embodiment of the present disclosure does not particularly limit how to obtain the target channel state information according to the plurality of third channel state information and the packet manner of packet pre-processing the initial channel state information in the target dimension.

[0116] In some embodiments, according to the plurality of third channel state information and a packet manner for packet pre-processing the initial channel state information in a target dimension, the step of obtaining target channel state information includes a step of sequentially combining the plurality of third channel state information in the target dimension to obtain target channel state information, and the plurality of first channel state information is obtained by dividing the resource granularity of each consecutive packet granularity number among the resource granularities in the target dimension of the initial channel state information into one resource granularity group.

[0117] In some examples, the terminal sequentially encodes and feeds back the first channel state information using a packet scheme that divides the resource granularity of each successive packet granularity into one resource granularity group, and the base station sequentially combines multiple pieces of third channel state information to obtain target channel state information with the same size as the initial channel state information.

[0118] In some other embodiments, the step of obtaining target channel state information in response to the plurality of third channel state information and a packet scheme for packet pre-processing the initial channel state information in a target dimension includes the steps of: determining a correspondence between a resource granularity in each piece of third channel state information and a resource granularity in the initial channel state information in response to the packet scheme for packet pre-processing the initial channel state information in the target dimension; and combining the resource granularity in the plurality of third channel state information in response to the correspondence between the resource granularity in each piece of third channel state information and the resource granularity in the initial channel state information to obtain target channel state information.

[0119] In some examples, the correspondence between the resource granularity in the third channel state information and the resource granularity in the initial channel state information includes that the resource granularity in the third channel state information corresponds to the resource granularity of the packet granularity number arranged at equal intervals in the initial channel state information, and the plurality of first channel state information are obtained by dividing the resource granularity of each of the packet granularity numbers arranged at equal intervals in the initial channel state information into one resource granularity group.

[0120] In some other examples, the correspondence between the resource granularity in the third channel state information and the resource granularity in the initial channel state information includes that the resource granularity in the third channel state information corresponds to some of the plurality of resource blocks in the initial channel state information, where the plurality of first channel state information are obtained by dividing the plurality of resource blocks in the initial channel state information into a plurality of resource granularity groups, the resource blocks include a plurality of resource granularities, and the sum of the number of resource granularities included in all the resource blocks divided into one resource granularity group is equal to the number of packet granularities.

[0121] In some embodiments, the resource granularity in the third channel state information corresponding to some of the plurality of resource blocks in the initial channel state information includes the resource granularity in the third channel state information corresponding to a resource granularity of a plurality of equally spaced resource blocks among the plurality of resource blocks in the initial channel state information, where the plurality of pieces of first channel state information are obtained by dividing the resource granularity of each of the plurality of equally spaced resource blocks among the plurality of resource blocks in the initial channel state information into one resource granularity group.

[0122] In some embodiments, the resource granularity in the third channel state information corresponding to some of the plurality of resource blocks in the initial channel state information includes the resource granularity in the third channel state information corresponding to a resource granularity of a plurality of resource blocks in a head-tail combination among the plurality of resource blocks in the initial channel state information, where the plurality of pieces of first channel state information are obtained by dividing the resource granularity of the plurality of resource blocks in the initial channel state information into a plurality of resource granularity groups according to a head-tail combination scheme.

[0123] In some other examples, the correspondence between the resource granularity in the third channel state information and the resource granularity in the initial channel state information includes that the resource granularity in the third channel state information corresponds to the resource granularity of the randomly arranged packet granularity numbers in the initial channel state information, and the plurality of first channel state information are obtained by randomly dividing the resource granularity of each packet granularity number in the initial channel state information into one resource granularity group.

[0124] In some embodiments, the step of obtaining target channel state information according to a plurality of third channel state information and a packet scheme for packet pre-processing the initial channel state information in a target dimension further includes a step of removing a filled number of redundant resource granularities in at least one piece of third channel state information when the number of resource granularities in the target dimension of the initial channel state information is not divisible by the number of packet granularities, and the redundant resource granularities correspond to the resource granularities filled in the at least one piece of first channel state information.

[0125] In addition, in an embodiment of the present disclosure, the number of filled redundant resource granularities characterized and in need of removal on the base station side is the same as the number of resource granularities characterized and in need of filling on the terminal side, waiting to be filled.

[0126] In some embodiments, removing redundant resource granularities in at least one third channel state information includes removing a filled number of resource granularities from the end in the third channel state information corresponding to a first target resource granularity group, the first target resource granularity group being a resource granularity group filled with the filled number of configured resource granularities.

[0127] In some other embodiments, removing redundant resource granularities in the at least one third channel state information includes removing a filled number of resource granularities from the end in the third channel state information corresponding to a first target resource granularity group, the first target resource granularity group being a resource granularity group filled by duplicating the filled number of first target resource granularities, and the first target resource granularity being one of the multiple resource granularities of the initial channel state information.

[0128] In some examples, the first target resource granularity is the penultimate resource granularity in the initial channel state information.

[0129] In some other embodiments, removing redundant resource granularities in at least one third channel state information includes removing a filled number of resource granularities from the end of the third channel state information corresponding to a first target resource granularity group, the first target resource granularity group being a resource granularity group filled by duplicating resource granularities in the waiting to be filled sequence, and the waiting to be filled sequence including the filled number of resource granularities of the initial channel state information.

[0130] In some examples, the waiting-to-fill sequence consists of a consecutive number of filled resource granularities for which the initial channel state information is in forward or reverse order starting from the second target resource granularity in the target dimension.

[0131] In some examples, the second target resource granularity is any one of the first resource granularity, the penultimate resource granularity, or the penultimate resource granularity in a target dimension of the initial channel state information.

[0132] In some other embodiments, removing the redundant resource granularity in the at least one third channel state information comprises removing the redundant resource granularity in the third channel state information corresponding to the at least one second target resource granularity group according to an overlapping filling position and an overlapping filling resource dimension, where the second target resource granularity group is the resource granularity group where the overlap and filling has been performed, the overlapping filling position characterizes a position of the second target resource granularity group, and the overlapping resource dimension characterizes a number of resource granularities filled in the second target resource granularity group.

[0133] In some examples, removing redundant resource granularities in the third channel state information corresponding to at least one second target resource granularity group according to the overlapping filling positions and the overlapping filling resource dimensions includes determining third channel state information corresponding to a first number of second target resource granularity groups, and removing a second number of resource granularities from the beginning in each third channel state information corresponding to the second target resource granularity group.

[0134] The first number of second target resource granularity groups are consecutive, and the first to second number of resource granularities in each second target resource granularity group are copies of the last to second number of resource granularity groups in a previous resource granularity group, respectively, and the product of the first number and the second number is equal to the filled number.

[0135] In some instances, the first number is equal to the filled number and the second number is one. In some examples, removing redundant resource granularities in the third channel state information corresponding to at least one second target resource granularity group according to the overlapping filling positions and the overlapping filling resource dimensions includes: determining, in the third channel state information corresponding to one second target resource granularity group, that a filled number of resource granularities have been filled in the second target resource granularity group; and removing the first filled number of resource granularities in the third channel state information corresponding to the second target resource granularity group.

[0136] In some examples, removing redundant resource granularity in the third channel state information corresponding to at least one second target resource granularity group according to the overlapping filling positions and the overlapping filling resource dimensions includes determining third channel state information corresponding to a plurality of second target resource granularity groups, where the overlapping resource dimensions of at least two second target resource granularity groups are different from each other, and removing redundant resource granularity in the third channel state information corresponding to each second target resource granularity group according to the overlapping resource dimensions of each second target resource granularity group.

[0137] In some embodiments, the step of obtaining target channel state information according to a plurality of third channel state information and a packet scheme for packet pre-processing the initial channel state information in a target dimension includes a step of combining the plurality of third channel state information according to a packet index to obtain the target channel state information.

[0138] The packet index characterizes the correspondence between a plurality of first channel state information and a plurality of resource granularities in the initial channel state information. In the embodiment of the present disclosure, the packet index may be determined by the base station at the same time when the base station determines the packet mode and transmits it to the terminal, or may be determined by the terminal and transmitted to the base station when the initial channel state information is packet pre-processed. The embodiment of the present disclosure is not particularly limited thereto.

[0139] In some embodiments, the target dimension includes at least one of a frequency domain dimension, a spatial domain dimension, a layer dimension, and a time domain dimension.

[0140] In some embodiments, if the target dimension includes a frequency domain dimension, the resource granularity includes a frequency domain granularity. If the target dimension includes a spatial domain dimension, the resource granularity includes an antenna port or a polarization direction. If the target dimension includes a layer dimension, the resource granularity includes a layer granularity. If the target dimension includes a time domain dimension, the resource granularity includes a time domain granularity.

[0141] In some embodiments, in S21, before receiving the second channel state information, the processing method further includes a step of transmitting packet signaling, where the packet signaling is used to instruct the terminal to packet pre-process the initial channel state information in a target dimension.

[0142] In some embodiments, the packet signaling includes a packet parameter, the packet parameter being: A packet identification bit is used to identify whether the initial channel state information needs to be packetized and processed; a divisibility identification bit used to identify whether the number of resource granularities in the target dimension of the initial channel state information is divisible by the number of packet granularities; Packet type information is used to indicate a packet type for packet pre-processing the initial channel state information; a packet granularity number used to indicate the number of resource granularities in a resource granularity group obtained by packetizing the initial channel state information; and resource granularity filling information used to indicate an overlap filling position and an overlap resource dimension of the second target resource granularity group when overlap filling.

[0143] In some embodiments, before receiving the second channel state information in S21, the processing method includes: configuring a second packet placement list, the second packet placement list including at least one packet scheme for packet pre-processing the initial channel state information; The method further includes the step of transmitting information of the second packet placement list.

[0144] In some embodiments, the method includes the steps of receiving a first packet placement list; The method further includes the step of generating packet signaling according to information of the first packet arrangement list.

[0145] In some embodiments, generating target channel state information in response to the plurality of third channel state information comprises receiving antenna polarization phase auxiliary information; and generating target channel state information in response to the polarization phase auxiliary information and the plurality of third channel state information.

[0146] In a third aspect, referring to FIG. 7 , some embodiments of the present disclosure provide a terminal, the terminal comprising: At least one processor 101; A memory 102 in which one or more programs are stored, the one or more programs being executed by at least one processor 101 to cause the at least one processor 101 to implement a method for processing any channel state information according to a first aspect of an embodiment of the present disclosure; A device connected between at least one processor 101 and a memory 102, At least one and at least one I / O interface 103 configured to enable information interaction between the processor 101 and the memory 102.

[0147] The processor 101 is a device having a data processing function, and includes, but is not limited to, a central processing unit (CPU). The memory 102 is a device having a data storage function, and includes, but is not limited to, a random access memory (RAM), a read-only memory (ROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory (FLASH). The RAM includes, for example, a synchronous dynamic random access memory (SDRAM) or a DDR (DDR SDRAM). The I / O interface (read A read / write interface 103 is connected between the processor 101 and the memory 102, and can realize information interaction between the processor 101 and the memory 102, and includes, but is not limited to, a data bus.

[0148] In some embodiments, the processor 101, memory 102 and I / O interface 103 are connected to each other via a bus 104, and thereby to other assemblies of the computing device.

[0149] In a fourth aspect, referring to FIG. 8 , some embodiments of the present disclosure provide a base station, the base station comprising: one or more processors 201; A memory 202 in which one or more programs are stored, the one or more programs being executed by one or more processors 201 to cause the one or more processors 201 to implement a channel state information processing method according to a second aspect of an embodiment of the present disclosure; and one or more I / O interfaces 203 coupled between the processor 201 and the memory 202 and configured to enable information interaction between the processor 201 and the memory 202.

[0150] The processor 201 is a device having a data processing function, including but not limited to a central processing unit. The memory 202 is a device having a data storage function, including but not limited to a random access memory (e.g., SDRAM, or DDR), a read-only memory, an electrically erasable programmable read-only memory, or a flash memory. The I / O interface (read / write interface) 203 is connected between the processor 201 and the memory 202, and can realize information interaction between the processor 201 and the memory 202, including but not limited to a data bus.

[0151] In some embodiments, the processor 201, memory 202 and I / O interface 203 are connected to each other via a bus 204, and thereby to other assemblies of the computing device.

[0152] In a fifth aspect, referring to FIG. 9 , some embodiments of the present disclosure provide a non-transitory computer-readable medium having a computer program stored thereon, the computer program being capable of, when executed by a processor, performing: At least one of the channel state information processing method according to a first aspect of the embodiment of the present disclosure and the channel state information processing method according to a second aspect of the embodiment of the present disclosure is realized.

[0153] In order to enable those skilled in the art to more clearly understand the technical solutions provided by the embodiments of the present disclosure, the technical solutions provided by the embodiments of the present disclosure are described in detail below through Examples 1 to 21.

[0154] Example 1 Example 1 is used to explain the channel state information format.

[0155]

number

[0156]

number

[0157]

number

[0158] In one embodiment, the initial channel state information needs to be packet pre-processed according to needs to generate first channel state information, and the process is shown in FIG.

[0159] Examples 2 to 6 are used to explain the frequency domain dimensional packet scheme.

[0160]

number

[0161]

number

[0162] Example 2

[0163]

number

[0164] Example 3 Example 3 aims to further improve performance by expanding the frequency domain resource span in each group.

[0165]

number

[0166] Example 4

[0167]

number

[0168] Example 5

[0169]

number

[0170] Example 6

[0171]

number

[0172] Examples 7 to 12 are used to explain solutions where the packets are not divisible.

[0173]

number

[0174] Example 7

[0175]

number

[0176] Example 8

[0177]

number

[0178] Example 9

[0179]

number

[0180] Example 10

[0181]

number

[0182] Example 11

[0183]

number

[0184] Example 12

[0185]

number

[0186] Examples 13 to 15 are used to explain the packet scheme in the spatial domain dimension.

[0187]

number

[0188] Example 13

[0189]

number

[0190] Example 14

[0191]

number

[0192] Example 15

[0193]

number

[0194] Example 16 Example 16 is used to explain the layer packet method.

[0195]

number

[0196]

number

[0197] Example 17 Example 17 is used to explain the time domain packet scheme.

[0198]

number

[0199]

number

[0200] Example 18 Example 18 is used to explain the encoder and decoder structures of the neural network.

[0201] The artificial intelligence can realize the feedback of the channel state information through a deep learning neural network. The neural network here is an autoencoder including at least an encoder and a decoder pair. Usually, the encoder is on the terminal side, and the decoder is on the base station side. The encoder usually includes two modules, a first processing module used to extract the features of the channel information, and a compression module used to compress the channel information. As shown in FIG. 24, the decoding module also includes two modules, a decompression module used to decompress the channel state information, and a second processing module used to recover the features of the decompressed channel state information.

[0202] The terminal obtains the first channel state information through packet pre-processing, and then transmits it to the encoder, and the first processing module extracts features of the channel information, and then outputs the second channel state information compressed through the compression module. The encoder includes a first processing module and a compression module. The first processing module is a sub-neural network, and includes C network layers. The network of each layer may be at least one of a fully connected layer, a convolution layer, a pooling layer, a direct connection layer, a batch normalization layer, etc. The compression module may include one fully connected layer, C1 convolution layers (or multiple residual blocks resblock, dense blocks denseblock), or a long and short term recording layer. The encoder may be a recurrent network such as a long short-term memory (LSTM), a gated recurrent neural network (GRU), or a Transformer structure. At the base station side, a decoder corresponding to the encoder includes a decompression module and a second processing module. The second processing module includes a processing module. The decoder receives the second channel state information, decompresses it, extracts features, and outputs third channel state information. The second processing module is also a sub-neural network, including L network layers. Here, the i-th layer may be a convolutional layer, or a pooling layer, a direct connection layer, or a combination thereof, or may be a residual network block, a dense block, which is composed of several convolutional layers. The decompression module may be a fully connected layer, C1 deconvolutional layers (e.g., an upsampling layer, an uppooling layer, a transposed convolutional layer, and their combinations with various residual blocks, dense blocks, and convolutional layers), LSTM, GUR, Transformer structure, etc. Exemplarily, the second channel state information may be quantized, coded, and modulated before being transmitted to the base station. The base station may dequantize, demodulate, and decode the channel state information through the received channel state information, and then use it as the input of the decoder, and the decoder outputs the third channel state information.

[0203] Example 19 Example 19 is used to illustrate the feedback polarization phase assistance scheme.

[0204] In the dual-polarized antenna packet feedback method, two antennas with different polarization directions have different polarization phases, so the polarization phase feedback of the antennas is extracted to obtain the chi- The recovery of channel information may be assisted. Suppose the initial channel information is as follows [Equation 1]:

[0205]

number

[0206]

number

[0207] The phase matrix is ​​given by the following Equation 3.

[0208]

number

[0209]

number

[0210] Example 20 Example 20 is used to explain how to obtain the parameters of the neural network.

[0211]

number

[0212] Example 21 Example 21 is used to explain how to set and obtain packet parameters.

[0213] The base station configures, sets and stores by itself according to the packet method in the above embodiment, and transmits the configured packet parameters to the terminal through higher layer and / or physical layer signaling. The terminal performs packet pre-processing on the acquired downlink channel state information according to the packet parameter signaling indication configured by the base station, and generates first channel state information.

[0214]

number

[0215] In one embodiment, the different dimension packet formats are arranged in a list. Illustratively, the frequency domain dimension packet is numbered 0, and the spatial domain dimension packet is numbered 1. packet is number 1, and the layer dimensionThe packets are numbered 2, etc., and are arranged as above. Depending on the number of the numbers, the numbers may be encoded using, but not limited to, a binary encoding (0, 1 bit) scheme. Correspondingly, the frequency domain dimension packet is numbered 00, and the spatial domain dimension packet is number 01, and the layer dimension The packet is number 10.

[0216] In one embodiment, the base station needs to set one bit to determine whether the resource packet is divisible (whether it needs to be filled so that it can be divisible). Bit 0 indicates that the packet resource does not need to be filled and is divisible, and bit 1 indicates that the packet resource is not divisible and needs to be filled. Of course, the contents represented by the 0 and 1 bits can be reversed.

[0217] In one embodiment, if the packet is not divisible, the base station may set it in a list according to the method of solving the problem of the packet not divisible in the above embodiment. By different methods, numbering from 0 and arranging as above. According to the number of numbers, the number may be encoded using, but not limited to, a binary encoding (0,1 bit) method. Correspondingly, numbering is set to 000 by the zero-filling method, the trailing channel information is duplicated as the number 001, and the trailing channel information is reversed and filled as the number 010.

[0218]

number

[0219] It should be noted that the number of bits encoded by the above embodiment may be binary adaptively encoded according to the total number of required resources. For example, if the indexes that need to be encoded are 3 in total, that is, only 2 binary bits need to be encoded.

[0220] The terminal feeds back the second channel information after the packet to the base station, and the third channel state information after the packet recovered by the base station selects final channel state information having the same dimension as the initial downlink channel state information as the final result fed back by the terminal according to the packet position information pre-arranged on the terminal side.

[0221] Similarly, the base station may configure signaling of a part of packet parameters, and the remaining part autonomously realizes packet pre-processing at the terminal side. In one embodiment, the terminal pre-configures and stores a packet configuration list, that is, a list of processing methods in which the above-mentioned packet resource is divisible and not divisible. Through upper layer and / or physical layer signaling, the terminal feeds back the packet configuration list information configured by the terminal to the base station. The base station needs to record and store the list to realize synchronization of the packet operation list between the terminal and the base station. At this time, the base station side only needs to configure a packet indication index, packet resource granularity number information, and an indication index of whether the packet is divisible or not, and send it to the terminal through upper layer and / or physical layer signaling. The terminal autonomously selects a corresponding solution method according to the relevant packet indication signaling sent by the base station, performs packet pre-processing on the acquired downlink channel state information, and generates first channel state information. The terminal needs to feed back to the base station the second channel state information processed by the encoder and the index / information of the packet processing method list selected by the terminal. The base station finally selects the third state information recovered by the decoder according to the index / information of the packet processing method list fed back by the terminal, and finally obtains the recovered downlink channel state information.Exemplarily, when the frequency domain resource channel is 32 subbands, the base station needs to pre-process the packet therefor, and each group is configured to be 5 subband channel information, and at this time, there is a situation where the packet is not divisible.If the method used by the terminal is to satisfy the packet divisibility by complementing the last three subband resource channels of the last group with zeros, the base station finally discards the last three subband channel information corresponding to the recovered third channel state information according to the position index in the last three subbands fed back by the terminal, thereby ensuring that the dimension is the same as the initial 32 subband channel information, and takes it as the downlink channel state information fed back by the final terminal.

[0222] In addition, if the packet is not divisible and the resource needs to be filled because the packet is divisible, the filled resource group is also compressed and fed back using the same network parameters. If the performance is deteriorated using the same network parameters, the terminal needs to select according to the instruction requirements fed back by the base station side, or the terminal needs to autonomously select other network parameters according to the instruction, and then compress and feed back.

[0223] Those skilled in the art can understand that all or some of the steps in the methods, the functional modules / units in the systems, the devices disclosed above can be implemented as software, firmware, hardware, and appropriate combinations thereof. In hardware embodiments, the division between the functional modules / units mentioned in the above description does not necessarily correspond to the division of physical assemblies. For example, one physical assembly may have multiple functions, or one function or step may be performed by several physical assemblies cooperatively. Some or all of the physical assemblies may be implemented as software executed by a processor, such as a central processing unit, a digital signal processor, or a microprocessor, or may be implemented as hardware, or may be implemented as an integrated circuit, such as an application specific integrated circuit. Such software may be distributed on a computer readable medium, which may include computer storage media (or non-transitory media) and communication media (or transitory media). As known to those skilled in the art, the term computer storage media includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (computer readable instructions, data structures, program modules, or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash or other memory technology, CD-ROM, digital versatile disks (DVDs) or other optical disks, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium which can be used to store the desired information and which can be accessed by a computer.Those skilled in the art also know that communication media typically include computer readable instructions, data structures, program modules or other data in a modulated data signal such as a carrier wave or other transport mechanism and may include any information delivery media.

[0224] In this specification, exemplary embodiments are disclosed, and specific terms are used, but they are generally used in an illustrative sense only and should not be construed as limiting. In some embodiments, unless otherwise specified, it will be apparent to one skilled in the art that features, characteristics, and / or elements described in combination with a particular embodiment may be used alone or in combination with features, characteristics, and / or elements described in combination with other embodiments. Thus, it will be apparent to one skilled in the art that various changes in form and details may be made without departing from the scope of the present disclosure as set forth in the claims.

[0225] It will be understood by those skilled in the art that the scope of the disclosure of the present invention is not limited to the specific examples described above, and that certain elements of the examples may be modified and substituted without departing from the spirit of the present application. The scope of the present application is limited by the appended claims.

Claims

1. Packet pre-processing the initial channel state information in a target dimension to obtain a plurality of first channel state information; encoding the plurality of first channel state information to generate second channel state information; feeding back the second channel state information; Including, How to process channel state information.

2. The step of packet pre-processing the initial channel state information in the target dimension to obtain the plurality of first channel state information includes: Dividing a plurality of resource granularities in the target dimension of the initial channel state information into a plurality of resource granularity groups to obtain the plurality of first channel state information, wherein the plurality of resource granularity groups correspond one-to-one to the plurality of first channel state information. The method of claim 1.

3. The step of dividing the plurality of resource granularities in the target dimension of the initial channel state information into the plurality of resource granularity groups to obtain the plurality of first channel state information includes: a step of dividing the resource granularities of each of the consecutive packet granularities among the plurality of resource granularities in the target dimension of the initial channel state information into one resource granularity group according to a packet granularity number, and obtaining the plurality of first channel state information, wherein the packet granularity number characterizes the number of resource granularities in the resource granularity group; The method of claim 2.

4. The step of dividing the plurality of resource granularities in the target dimension of the initial channel state information into the plurality of resource granularity groups to obtain the plurality of first channel state information includes: A step of dividing the resource granularities of the packet granularity number arranged at equal intervals among the plurality of resource granularities in the target dimension of the initial channel state information into one resource granularity group according to a packet granularity number, and obtaining the plurality of first channel state information, wherein the packet granularity number characterizes the number of resource granularities in the resource granularity group. The method of claim 2.

5. The step of dividing the plurality of resource granularities in the target dimension of the initial channel state information into the plurality of resource granularity groups to obtain the plurality of first channel state information includes: a step of dividing a plurality of resource blocks in the initial channel state information into a plurality of resource granularity groups according to a packet granularity number to obtain the plurality of first channel state information, the packet granularity number characterizing a number of resource granularities in a resource granularity group, a resource block including a portion of the plurality of resource granularities, and a sum of the number of resource granularities included in all resource blocks divided into one of the resource granularity groups being equal to the packet granularity number; The method of claim 2.

6. The step of dividing the plurality of resource blocks in the initial channel state information into the plurality of resource granularity groups according to the packet granularity number and obtaining the plurality of first channel state information includes: According to a packet granularity number, the resource granularity of each of the plurality of resource blocks arranged at equal intervals among the plurality of resource blocks in the initial channel state information is divided into one of the resource granularity groups, and the plurality of first channel state information is obtained. The method of claim 5.

7. The step of dividing the plurality of resource blocks in the initial channel state information into the plurality of resource granularity groups according to the packet granularity number and obtaining the plurality of first channel state information includes: Dividing the resource granularity in the plurality of resource blocks into the plurality of resource granularity groups in a head and tail combination manner according to a packet granularity number, and obtaining the plurality of first channel state information; The method of claim 5.

8. The step of dividing the plurality of resource granularities in the target dimension of the initial channel state information into the plurality of resource granularity groups to obtain the plurality of first channel state information includes: A step of randomly dividing the resource granularities of each of the packet granularity numbers among the plurality of resource granularities in the target dimension of the initial channel state information into one of the resource granularity groups according to a packet granularity number, and obtaining the plurality of first channel state information, wherein the packet granularity number characterizes the number of resource granularities in the resource granularity group. The method of claim 2.

9. The step of dividing the plurality of resource granularities in the target dimension of the initial channel state information into the plurality of resource granularity groups to obtain the plurality of first channel state information includes: and filling resource granularities when the number of resource granularities in the target dimension of the initial channel state information is not divisible by a packet granularity number, the packet granularity number characterizing a number of resource granularities in the resource granularity group.

9. The method according to claim 2 .

10. The step of filling the resource granularity includes: a step of filling a first target resource granularity group with a set resource granularity of the number of waiting to be filled according to the number of waiting to be filled, the number of waiting to be filled being the number of resource granularities that need to be filled, and the first target resource granularity group being a resource granularity group containing a number of resource granularities smaller than the number of packet granularities before filling the resource granularities; The method of claim 9.

11. The step of filling the resource granularity includes: a step of duplicating a first target resource granularity by the number of waiting to be filled according to the number of waiting to be filled, and filling the first target resource granularity group, the first target resource granularity being one of the plurality of resource granularities in the target dimension of the initial channel state information, the number of waiting to be filled being the number of resource granularities that need to be filled, and the first target resource granularity group being a resource granularity group including a number of resource granularities smaller than the number of packet granularities before filling the resource granularities; The method of claim 9.

12. The step of filling the resource granularity includes: a step of duplicating resource granularities in a sequence waiting to be filled according to a number of waiting to be filled and filling the first target resource granularity group, the sequence waiting to be filled including the resource granularities of the number of waiting to be filled, the number of waiting to be filled being a number of resource granularities that need to be filled, and the first target resource granularity group being a resource granularity group including a number of resource granularities smaller than the number of packet granularities before filling the resource granularities; The method of claim 9.

13. The step of duplicating the resource granularity in the sequence of resources waiting to be filled according to the number of resources waiting to be filled and filling the first target resource granularity group includes: A step of configuring the resource granularities of the number of waiting to be filled successively in a forward or reverse order starting from a second target resource granularity as the waiting to be filled sequence; replicating the resource granularity in the sequence waiting to be filled to fill the first target resource granularity group; Including, The method of claim 12.

14. The step of filling the resource granularity includes: determining an overlapping filling position and an overlapping resource size of at least one second target resource granularity group as a function of the number of waiting to be filled, the overlapping filling position characterizing a position of the second target resource granularity group and the overlapping resource size characterizing a number of resource granularities in the second target resource granularity group that need to be filled; overlapping and filling each of the second target resource granularity groups according to an overlapping filling position and an overlapping resource dimension of each of the second target resource granularity groups; Including, The method of claim 9.

15. The step of determining an overlapping filling position and an overlapping resource size of at least one second target resource granularity group according to the number of waiting to be filled includes: determining a first number of consecutive second target resource granularity groups according to the number of waiting to be filled, each of the second target resource granularity groups having an overlapping resource dimension equal to a second number, and a product of the first number and the second number being equal to the number of waiting to be filled; The step of overlapping and filling each of the second target resource granularity groups according to an overlapping filling position and an overlapping resource dimension of each of the second target resource granularity groups includes: a step of duplicating the last second number of resource granularities in a resource granularity group preceding each of the second target resource granularity groups to be a first second number of resource granularities in each of the second target resource granularity groups; 15. The method of claim 14.

16. The step of determining an overlapping filling position and an overlapping resource size of at least one second target resource granularity group according to the number of waiting to be filled includes: determining a second target resource granularity group, the overlapping resource dimension of the second target resource granularity group being the number of waiting to be filled; The step of overlapping and filling each of the second target resource granularity groups according to an overlapping filling position and an overlapping resource dimension of each of the second target resource granularity groups includes: A step of duplicating the resource granularity of the last number of waiting to be filled in one resource granularity group before the second target resource granularity group according to the number of waiting to be filled, and setting the resource granularity of the first number of waiting to be filled in the second target resource granularity group.

15. The method of claim 14.

17. The step of determining an overlapping filling position and an overlapping resource size of at least one second target resource granularity group according to the number of waiting to be filled includes: determining a plurality of second target resource granularity groups and an overlapping resource dimension of each of the second target resource granularity groups, wherein the overlapping resource dimensions of at least two of the second target resource granularity groups are different from each other; 15. The method of claim 14.

18. The target dimension includes at least one of a frequency domain dimension, a spatial domain dimension, a layer dimension, and a time domain dimension.

18. The method of claim 1 .

19. Before packet pre-processing the initial channel state information in the target dimension to obtain the plurality of first channel state information, the processing method includes: The method further includes receiving packet signaling, the packet signaling including packet parameters; The step of packet pre-processing the initial channel state information in the target dimension to obtain the plurality of first channel state information includes: packet pre-processing the initial channel state information in the target dimension according to the packet parameters to obtain the plurality of first channel state information; 19. The method of claim 1 .

20. receiving second channel state information; decoding the second channel state information to obtain a plurality of third channel state information, the plurality of third channel state information corresponding one-to-one to a plurality of first channel state information obtained by packet pre-processing the initial channel state information in a target dimension; generating target channel state information in response to the plurality of third channel state information; Including, How to process channel state information.

21. The step of generating the target channel state information in response to the plurality of third channel state information includes: obtaining the target channel state information according to the plurality of third channel state information and a packet manner for packet pre-processing the initial channel state information in the target dimension; 21. The method of claim 20.

22. The step of obtaining the target channel state information according to the plurality of third channel state information and a packet manner for packet pre-processing the initial channel state information in the target dimension includes: A step of sequentially combining the plurality of third channel state information in the target dimension to obtain the target channel state information, wherein the plurality of first channel state information are obtained by dividing the resource granularity of each consecutive packet granularity number among the plurality of resource granularities in the target dimension of the initial channel state information into one resource granularity group.

22. The method of claim 21.

23. The step of obtaining the target channel state information according to the plurality of third channel state information and a packet manner for packet pre-processing the initial channel state information in the target dimension includes: determining a correspondence relationship between a resource granularity in each third channel state information and a resource granularity in the initial channel state information according to a packet manner for packet pre-processing the initial channel state information in the target dimension; According to a correspondence relationship between a resource granularity in each of the third channel state information and a resource granularity in the initial channel state information, a resource granularity in each of the third channel state information is combined to obtain the target channel state information; Including, 22. The method of claim 21.

24. a correspondence relationship between a resource granularity in the third channel state information and a resource granularity in the initial channel state information includes that the resource granularity in the third channel state information corresponds to a resource granularity of a packet granularity number arranged at equal intervals in the initial channel state information, and the plurality of pieces of first channel state information are obtained by dividing the resource granularity of each of the packet granularity numbers arranged at equal intervals in the initial channel state information into one resource granularity group.

24. The method of claim 23.

25. A correspondence relationship between a resource granularity in the third channel state information and a resource granularity in the initial channel state information includes that the resource granularity in the third channel state information corresponds to some of a plurality of resource blocks in the initial channel state information; the plurality of first channel state information are obtained by dividing the plurality of resource blocks in the initial channel state information into a plurality of resource granularity groups, each resource block includes a plurality of resource granularities, and a sum of the number of resource granularities included in all the resource blocks divided into one of the resource granularity groups is equal to a packet granularity number; 24. The method of claim 23.

26. a correspondence relationship between a resource granularity in the third channel state information and a resource granularity in the initial channel state information includes that the resource granularity in the third channel state information corresponds to a resource granularity of a packet granularity number randomly arranged in the initial channel state information, and the plurality of pieces of first channel state information are obtained by randomly dividing the resource granularity of each packet granularity number in the initial channel state information into one resource granularity group.

24. The method of claim 23.

27. The step of obtaining the target channel state information according to the plurality of third channel state information and a packet manner for packet pre-processing the initial channel state information in the target dimension includes: The method further includes the step of removing a filled number of redundant resource granularities in at least one third channel state information when the number of resource granularities in the target dimension of the initial channel state information is not divisible by the number of packet granularities, the redundant resource granularities corresponding to the resource granularities filled in the at least one first channel state information.

27. A method according to any one of claims 22 to 26.

28. The step of removing redundant resource granularity in the at least one third channel state information comprises: removing redundant resource granularities in the third channel state information corresponding to at least one second target resource granularity group according to an overlapping filling position and an overlapping filling resource dimension, the second target resource granularity group being a resource granularity group where overlapping and filling has been performed, the overlapping filling position characterizing a position of the second target resource granularity group, and the overlapping resource dimension characterizing a number of resource granularities to be filled into the second target resource granularity group; 28. The method of claim 27.

29. The target dimension includes at least one of a frequency domain dimension, a spatial domain dimension, a layer dimension, and a time domain dimension.

29. A method according to any one of claims 20 to 28.

30. Before receiving the second channel state information, the processing method includes: The method further includes the step of transmitting packet signaling, the packet signaling being used to instruct a terminal to packet pre-process the initial channel state information in a target dimension.

30. The method of claim 20.

31. At least one processor; a memory in which one or more programs are stored, the one or more programs being executed by the at least one processor, causing the at least one processor to implement the method for processing channel state information according to any one of claims 1 to 19; At least one I / O interface coupled between the processor and the memory and configured to provide information interaction between the processor and the memory; Including, Terminal.

32. At least one processor; a memory in which one or more programs are stored, the one or more programs being executed by the at least one processor, causing the at least one processor to implement the method for processing channel state information according to any one of claims 20 to 30; At least one I / O interface coupled between the processor and the memory and configured to provide information interaction between the processor and the memory; Including, Base station.

33. A computer-readable medium having a computer program stored thereon, the program, when executed by a processor, A method for processing channel state information according to any one of claims 1 to 19; A method for processing channel state information according to any one of claims 20 to 30; To achieve at least one of the following: Computer-readable medium.

Images