Method for constructing downlink codebook

Abstract

The embodiment of the invention provides a method for constructing a downlink codebook. The method comprises the following steps: acquiring uplink channel measurement information of current user equipment (UE) corresponding to a precoding matrix indicator (PMI) codebook and downlink channel measurement information of the current UE; inputting the uplink channel measurement information and the downlink channel measurement information into a pre-trained PMI enhancement model to obtain a PMI enhancement weight output by the PMI enhancement model; and according to the PMI enhancement weight, a PMI codebook is constructed, and the constructed PMI codebook is used for scheduling downlink channel data of the current UE. The embodiment of the invention at least solves the problem that the performance gain of SU-MIMO and MU-MIMO cannot be fully reflected and the spectrum efficiency and throughput of a physical cell are affected due to the fact that quantization precision loss of different degrees exists in each PMI codebook type in the related technology.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of communications, and in particular to a method for constructing a downlink codebook. Background Art

[0002] In frequency-division duplex (FDD) wireless communication systems, due to the non-reciprocal nature of uplink and downlink channels, when a physical cell performs frequency-division or space-division scheduling for an accessing user, the downlink precoding matrix indicator (PMI) weight must rely solely on the PMI measured and fed back by the user's Channel State Information-Reference Signal (CSI-RS). However, each PMI codebook type suffers from varying degrees of quantization accuracy loss, which prevents the full performance gains of single-user multiple-input multiple-output (SU-MIMO) and multi-user multiple-input multiple-output (MU-MIMO) technologies from being realized, impacting the physical cell's spectrum efficiency and throughput. Summary of the Invention

[0003] The embodiments of the present application provide a method and apparatus for constructing a downlink codebook to at least address the problem in the related art that various PMI codebook types suffer from varying degrees of quantization accuracy loss, resulting in the inability to fully reflect the performance gains of SU-MIMO and MU-MIMO, thereby affecting the spectrum efficiency and throughput of the physical cell.

[0004] According to one embodiment of the present application, a method for constructing a downlink codebook is provided, including:

[0005] Obtaining uplink channel measurement information of a current user equipment UE and downlink channel measurement information of the current UE corresponding to a precoding matrix indication PMI codebook;

[0006] Inputting the uplink channel measurement information and the downlink channel measurement information into a pre-trained PMI enhancement model to obtain a PMI enhancement weight output by the PMI enhancement model;

[0007] The PMI codebook is constructed according to the PMI enhancement weight, wherein the constructed PMI codebook is used to schedule downlink channel data of the current UE.

[0008] According to another embodiment of the present application, a computer-readable storage medium is provided, in which a computer program is stored. The computer program is configured to execute the steps of any one of the above method embodiments when run.

[0009] According to another embodiment of the present application, an electronic device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to perform the steps in any one of the above method embodiments.

[0010] According to another embodiment of the present application, a computer program product is also provided, including a computer program, wherein the computer program is configured by a processor to run the computer program to execute the steps in any of the above method embodiments. The present application obtains the downlink channel measurement information of the current user equipment UE and the uplink channel measurement information of the current UE corresponding to the precoding matrix indication PMI codebook; inputs the downlink channel measurement information and the uplink channel measurement information into a pre-trained PMI enhancement model to obtain the PMI enhancement weight output by the PMI enhancement model; constructs a PMI codebook according to the PMI enhancement weight, wherein the constructed PMI codebook is used to schedule the downlink channel data of the current UE, which can optimize the downlink PMI shaping gain of users in the cell, thereby achieving the effect of improving the cell throughput and improving the user's downlink experience. It solves the problem in the related art that each PMI codebook type has different degrees of quantization accuracy loss, resulting in the inability to fully reflect the performance gain of SU-MIMO and MU-MIMO, affecting the spectrum efficiency and throughput of the physical cell. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 This is a hardware structure block diagram of a mobile terminal for a method of constructing a downlink codebook according to an embodiment of the present application;

[0012] Figure 2 is a flowchart of a method for constructing a downlink codebook according to an embodiment of the present application;

[0013] Figure 3 It is a schematic diagram of the model training topology structure according to an embodiment of the present application. DETAILED DESCRIPTION

[0014] The embodiments of the present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0015] It should be noted that the terms "first", "second", etc. in the description and claims of this application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence.

[0016] The base station side can usually configure different types of PMI codebooks for the UE, such as Type1, Type2-R15, eType2-R16, FeType2-R17, etc. Taking the Type1 codebook as an example, the existing technology usually uses the base station to construct the downlink service transmission weight based on the PMI measured and fed back by the user, in conjunction with the interference suppression processing process. However, since the Type1 codebook only fits the strongest main path in the wireless channel and the base station side cannot accurately know the downlink channel status, it cannot effectively compensate for it, resulting in a significant loss of shaping gain and accuracy, affecting cell throughput; especially in scenarios with non-line-of-sight (NLos), dense user distribution, and wireless channels subject to large external disturbances, the use of such a weight construction method not only fails to effectively suppress interference, but is also prone to problems such as increased shaping accuracy deviation and poor user experience.

[0017] Based on the above technical problems, an embodiment of the present application proposes a method for constructing a downlink codebook, which constructs the PMI original weight of the PMI codebook based on the PMI codebook pre-configured for the UE, and then enhances the PMI original weight based on the PMI original weight, the obtained UE downlink channel measurement information and the UE uplink channel measurement information. Based on the enhanced PMI original weight, the PMI codebook is reconstructed to perform frequency division scheduling or space division scheduling according to the reconstructed PMI codebook, thereby enhancing the shaping effect and accuracy, thereby achieving the purpose of improving cell throughput and improving user downlink experience, and solving problems such as increased shaping accuracy deviation and poor user experience.

[0018] The method embodiments provided in the embodiments of the present application can be executed in a mobile terminal, a computer terminal or a similar computing device. Taking running on a mobile terminal as an example, Figure 1 1 is a hardware structure block diagram of a mobile terminal for a method of constructing a downlink codebook according to an embodiment of the present application. Figure 1 As shown, the mobile terminal may include one or more ( Figure 1 Only one is shown) a processor 102 (the processor 102 may include but is not limited to a microprocessor MCU or a programmable logic device FPGA and other processing devices) and a memory 104 for storing data, wherein the mobile terminal may also include a transmission device 106 and an input and output device 108 for communication functions. It will be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the mobile terminal. Figure 1 More or fewer components than shown, or with Figure 1 Different configurations shown.

[0019] The memory 104 can be used to store computer programs, for example, software programs and modules of application software, such as the computer program corresponding to the method for constructing a downlink codebook in the embodiment of the present application. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, that is, implementing the above method. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include a memory remotely located relative to the processor 102, and these remote memories may be connected to the mobile terminal via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0020] The transmission device 106 is used to receive or send data via a network. A specific example of the aforementioned network may include a wireless network provided by the mobile terminal's communications provider. In one embodiment, the transmission device 106 includes a network interface controller (NIC), which can be connected to other network devices via a base station to enable communication with the Internet. In another embodiment, the transmission device 106 may be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0021] In this embodiment, a method for constructing a downlink codebook running on the above mobile terminal or network architecture is provided. Figure 2 is a flowchart of a method for constructing a downlink codebook according to an embodiment of the present application. Figure 2 As shown, the process includes the following steps:

[0022] Step S201: obtaining uplink channel measurement information of a current user equipment UE and downlink channel measurement information of a current UE corresponding to a precoding matrix indication PMI codebook.

[0023] Exemplarily, the method of the embodiment of the present application can be applied to a base station, and the corresponding current user equipment UE uplink channel measurement information, current UE downlink channel measurement information, etc. can be obtained through the base station.

[0024] In an exemplary embodiment, the downlink channel measurement information of the UE includes an original PMI weight, and the method further includes:

[0025] According to a precoding matrix indication PMI codebook pre-configured for the UE, a channel state information CSI code stream reported by the UE is parsed; and according to the CSI code stream, a PMI original weight of the PMI codebook is constructed.

[0026] For example, the base station can parse the CSI code stream reported by the UE according to the codebook type configured for the UE, such as Type1, Type2-R15, eType2-R16 or FeType2-R17, and reconstruct the original PMI weight W of the corresponding downlink PMI codebook under the premise of following the protocol process. 原始PMI .

[0027] In an exemplary embodiment, the UE downlink channel measurement information also includes at least one of the following: PMI, rank indicator RI, channel quality indication CQI, and reference channel received power RSRP reported by the UE; the UE uplink channel measurement information includes at least one of the following: uplink channel projection energy calculated by the user uplink sounding reference signal SRS, multipath arrival angle direction DOA, channel angle spread, multipath delay, signal to interference plus noise ratio SRS SINR of the sounding reference signal, and UE moving speed.

[0028] For example, the downlink channel measurement information may include but is not limited to UE-reported PMI, RI, channel quality indication (CQI), reference signal received power (RSRP) and other characteristic quantities obtained based on the cell downlink CSI-RS measurement signal; the uplink channel measurement information may include but is not limited to the uplink channel (SRSH) projection energy calculated based on the user uplink sounding reference signal (SRS), the direction of arrival (DOA) of the multipath radiation signal, the channel angle spread, the multipath delay, the SRS SINR and the UE moving speed and other characteristic quantities.

[0029] The base station may construct an original PMI weight according to the UE-reported PMI in the downlink channel measurement information.

[0030] Step S202: Input the downlink channel measurement information and the uplink channel measurement information into a pre-trained PMI enhancement model to obtain a PMI enhancement weight output by the PMI enhancement model.

[0031] For example, the original PMI weight W 原始PMI , other downlink channel measurement information corresponding to the PMI codebook except the original PMI weight, and uplink channel measurement information are used as the input of the pre-trained PMI enhancement model to obtain the PMI enhancement weight W output by the PMI enhancement model 增强PMI .

[0032] As an example, the pre-trained PMI enhancement model can extract the W of the above input PMI codebook respectively.原始PMI , the real and imaginary parts of other downlink channel measurement information and uplink channel measurement information corresponding to the PMI codebook except the original PMI weight, perform feature learning on the real and imaginary parts respectively, and output the enhanced downlink PMI codebook W 增强PMI The real part (W 增强PMI，real ), imaginary part (W 增强PMI，imag ).

[0033] For example, W 增强PMI =W 增强PMI，real +j·W 增强PMI，imag

[0034] Step S203: construct the PMI codebook according to the PMI enhancement weight, wherein the constructed PMI codebook is used to schedule downlink channel data of the current UE.

[0035] As an example, a codebook can be reconstructed according to the real part and imaginary part of the enhanced PMI weight to schedule the downlink channel data of the UE to the UE according to the reconstructed PMI codebook, for example, frequency division scheduling or space division scheduling is performed on the user based on the PMI enhancement weight in the reconstructed PMI codebook.

[0036] For example, Figure 3 This is a schematic diagram of the model training topology structure according to an embodiment of the present application. Figure 3 As shown, the base station side can obtain uplink signals and obtain uplink channel measurement information based on the uplink signals (for example, uplink SRS signals); the base station side can record the information of the scheduling process through SPA log and print the scheduling results statistically.

[0037] The UE may obtain UE downlink frequency domain data based on the downlink channel measurement information.

[0038] The base station side can obtain the UE downlink frequency domain data, and determine the type1 codebook and the PMI weight of the type1 codebook corresponding to the UE downlink frequency domain data based on the UE downlink frequency domain data; the downlink CSI channel H can be extracted based on the UE downlink frequency domain data, and the ideal precoding V of the downlink channel H can be obtained based on the downlink CSI channel H.

[0039] The base station side can supervise and train the PMI enhancement model based on the PMI weight of the type1 codebook, uplink channel measurement information, UE downlink frequency domain data other than the PMI weight of the type1 codebook, and the ideal precoding V of the downlink channel H as a constraint.

[0040] The base station side can input the PMI weight of the type 1 codebook and the uplink channel measurement information into the trained PMI enhancement model to obtain the enhanced PMI weight W output by the trained PMI enhancement model. 增强PMI , can be based on W增强PMI Reconstruct the codebook.

[0041] In an exemplary embodiment, the current UE is a single UE, and the method further includes:

[0042] According to the timeliness condition and the principle of maximizing spectrum efficiency, the PMI original weight or the PMI enhanced weight is selected from the PMI codebook, and single UE downlink channel data scheduling is performed according to the PMI original weight or the PMI enhanced weight.

[0043] For example, in a single-user multiple-input and multiple-output (SU-MIMO) scenario, that is, when the current UE is a single UE, a PMI original weight or a PMI enhanced weight may be selected from a PMI codebook to perform frequency division scheduling.

[0044] As an example, according to the timeliness condition and the principle of maximizing spectrum efficiency, the original PMI weight or the enhanced PMI weight can be selected from the PMI codebook, and single UE downlink channel data scheduling can be performed based on the original PMI weight or the enhanced PMI weight.

[0045] In an exemplary embodiment, the current UE is a multi-UE, and the method further includes:

[0046] In response to the multiple UEs satisfying the UE empty allocation pair condition, selecting the PMI original weight or the PMI enhanced weight of the paired UE from the PMI codebook for user downlink channel data scheduling according to the timeliness condition and the principle of minimizing interference between paired UEs; or,

[0047] In response to the multiple UEs not satisfying the UE empty allocation pair condition, according to the timeliness condition and the principle of maximizing spectrum efficiency, the PMI original weights or the PMI enhanced weights corresponding to the multiple UEs are respectively selected from the PMI codebook, and multi-UE downlink channel data scheduling is performed according to the PMI original weights or the PMI enhanced weights.

[0048] For example, in a multi-user multiple-input and multiple-output MU-MIMO scenario, that is, when the current UE is multiple UEs, the original PMI weight or the enhanced PMI weight can be selected from the PMI codebook to perform space division scheduling or frequency division scheduling.

[0049] As an example, when multiple UEs meet the UE empty allocation pair condition, the original PMI weight or the enhanced PMI weight of the paired UE can be selected from the PMI codebook for user downlink channel data scheduling based on the timeliness condition and the principle of minimizing interference between paired UEs.

[0050] If multiple UEs do not meet the UE empty allocation pair condition, the PMI original weights or PMI enhanced weights corresponding to the multiple UEs can be selected from the PMI codebook according to the timeliness condition and the principle of maximizing spectrum efficiency, and multi-UE downlink channel data scheduling can be performed according to the PMI original weights or the PMI enhanced weights.

[0051] The above steps address the issue in related technologies where various PMI codebook types suffer varying degrees of quantization accuracy loss, resulting in the inability to fully reflect the performance gains of SU-MIMO and MU-MIMO, impacting the spectrum efficiency and throughput of physical cells. This improves beamforming gain and accuracy, thereby improving cell throughput and spectrum efficiency.

[0052] The method of the embodiment of the present application can be applied to all codebook types of the current NR protocol evolution. The method can be quickly deployed and can adaptively learn and match application scenarios online according to the differences in channel characteristics of physical cells.

[0053] In an exemplary embodiment, the PMI enhancement model is obtained by:

[0054] Obtain the training data set for the PMI enhancement model to be trained;

[0055] Setting the constraints and solution objectives of the PMI enhancement model to be trained;

[0056] The PMI enhancement model to be trained is trained according to the training data set, the constraint conditions and the solution goal to obtain the PMI enhancement model.

[0057] Exemplarily, a training data set of the PMI enhancement model to be trained can be obtained, and constraints and solution objectives of the PMI enhancement model to be trained can be set. The PMI enhancement model to be trained can be trained according to the training data set, constraints and solution objectives to obtain a trained PMI enhancement model.

[0058] In an exemplary embodiment, the training of the to-be-trained PMI enhancement model according to the training data set, the constraint conditions, and the solution goal to obtain the PMI enhancement model includes:

[0059] According to the constraints and the solution goals, the historical UE downlink channel measurement information and the historical UE uplink channel measurement information are used as inputs of the PMI enhancement model to be trained, and the target PMI weight is used as output of the PMI enhancement model to be trained, so as to obtain the PMI enhancement model.

[0060] In an exemplary embodiment, the downlink channel includes multiple data layers, wherein the constraints include: orthogonalization of the PMI enhancement weights of the multiple data layers output by the PMI enhancement model; and the solution objective includes: maximizing the correlation between the PMI enhancement weights output by the PMI enhancement model and the target PMI weights.

[0061] As an example, the PMI enhancement model to be trained can be trained by using the training data set of the PMI enhancement model to be trained, the orthogonalization of the PMI enhancement weights of multiple data layers output by the PMI enhancement model as a constraint condition, and the maximization of the correlation between the PMI enhancement weights output by the PMI enhancement model and the target PMI weights as the solution goal.

[0062] For example, we can build an optimization model with the goal of maximizing the correlation between the PMI weight output by the model and the ideal precoding V of the downlink channel, and with the orthogonalization of the multi-stream PMI layers output by the model as a constraint:

[0063]

[0064]

[0065] As an example, a training dataset may be used to optimize the PMI enhancement model to be trained based on constraints.

[0066] As an example, optimization methods include but are not limited to traditional convex optimization theory modeling and PMI enhancement model learning, while the PMI enhancement models to be trained include but are not limited to decision tree models, recurrent neural networks (RNNs), convolutional neural networks (CNNs), etc. Model training methods include but are not limited to supervised learning, unsupervised learning, reinforcement learning, etc. The ideal precoding V of the downlink channel H can be obtained by periodically triggering user reports on the base station side.

[0067] In an exemplary embodiment, before obtaining a training data set for a PMI enhancement model to be trained, the method further includes:

[0068] Obtain historical UE downlink channel measurement information, historical UE uplink channel measurement information and target PMI weight;

[0069] A training data set for the to-be-trained PMI enhancement model is constructed according to the historical UE downlink channel measurement information, the historical UE uplink channel measurement information, and the target PMI weight.

[0070] Exemplarily, historical UE downlink channel measurement information, historical UE uplink channel measurement information and target PMI weights may be obtained, and a training data set for the PMI enhancement model to be trained may be constructed based on the historical UE downlink channel measurement information, historical UE uplink channel measurement information and target PMI weights.

[0071] In an exemplary embodiment, the historical UE downlink channel measurement information includes at least one of the following: PMI, rank index RI, channel quality indication CQI, and reference signal received power RSRP reported by the UE; the historical UE uplink channel measurement information includes at least one of the following: uplink channel projection energy calculated by the user uplink sounding reference signal SRS, multipath arrival angle direction DOA, channel angle spread, multipath delay, signal to interference plus noise ratio SRS SINR of the sounding reference signal, and UE moving speed.

[0072] As an example, before obtaining a training data set for the PMI enhancement model to be trained, uplink channel measurement information, downlink channel measurement information, neighboring cell interference information, historical scheduling information, etc. of users in the cell may be obtained. The uplink channel measurement information may include, but is not limited to, feature quantities such as uplink channel (SRSH) projection energy calculated based on the user's uplink sounding reference signal SRS, multipath DOA, channel angle spread, multipath delay, SRS SINR, and mobile speed; the downlink channel measurement information may include, but is not limited to, feature quantities such as UE-reported PMI, RI, CQI, and reference signal received power RSRP obtained based on the cell's downlink CSI-RS measurement signal.

[0073] As an example, the necessary feature quantities for model training may include UE reported PMI, uplink channel (SRSH) projection energy, multipath DOA, channel angle spread, multipath delay, CQI, reference channel received power RSRP, etc.

[0074] As an example, the acquired feature quantity information and the target PMI weight may be combined to construct a training data set for the PMI enhancement model to be trained.

[0075] In an exemplary embodiment, the historical UE downlink channel measurement information includes a UE-reported PMI, and constructing a training data set for the PMI enhancement model to be trained based on the historical UE downlink channel measurement information, the historical UE uplink channel measurement information, and the target PMI weight includes:

[0076] Selecting, according to the time corresponding to the UE reporting the PMI, historical UE uplink channel measurement information whose measurement time is close to the time corresponding to the UE reporting the PMI;

[0077] The historical UE downlink channel measurement information, the historical UE uplink channel measurement information whose measurement time is close to the time corresponding to the UE reporting the PMI, and the target PMI weight are combined to construct a training data set for the PMI enhancement model to be trained.

[0078] As an example, according to the time corresponding to the UE reporting PMI, a set of uplink channel measurement information closest to the SRS measurement time on the base station side can be screened, including but not limited to uplink channel (SRS H) projection energy, multipath DOA, channel angle extension, multipath delay, etc. The screened uplink channel measurement information, downlink channel measurement information and the ideal precoding V of the downlink channel H corresponding to the reported PMI can be combined to construct a training data set.

[0079] The method for constructing a downlink codebook according to the embodiment of the present application is further illustrated below through two examples.

[0080] Example 1

[0081] Taking the SU-MIMO system as an example, the following steps may be specifically included:

[0082] Step 1: The base station side obtains the downlink channel measurement information of the user UE to be scheduled, such as the CSI-RS reference signal measurement result. The downlink channel measurement information may include but is not limited to the codebook index PMI1, PMI2, RI, CQI, reference signal received power RSRP, etc.

[0083] Step 2: The base station obtains uplink channel measurement information of the user UE to be scheduled, such as the SRS reference signal measurement result. The uplink channel measurement information may include but is not limited to uplink channel (SRS H) projection energy, multipath DOA, channel angle spread, multipath delay, SRS SINR, and mobile speed;

[0084] Step 3: The base station inputs the acquired UE uplink channel measurement information and downlink channel measurement information into the PMI enhancement model;

[0085] Step 4: The PMI enhancement model can perform feature learning based on the input UE uplink channel measurement information and downlink channel measurement information, and output the enhanced downlink PMI codebook W 增强PMI ;

[0086] Step 5: The base station side can dynamically select W based on the enhanced downlink PMI codebook based on the timeliness condition and the principle of maximizing spectrum efficiency. 原始PMI or W 增强PMI Physical Downlink Shared Channel (PDSCH) scheduling is performed to improve the downlink experience rate of a single user.

[0087] Example 2

[0088] Taking the MU-MIMO system as an example, assuming that UEs are spatially divided and the users to be spatially divided include UE1 and UE2, the following steps may be specifically performed:

[0089] Step 1: The base station side obtains the downlink channel measurement information of the users UE1 and UE2 to be space-divided, such as the CSI-RS reference signal measurement results. The downlink channel measurement information may include but is not limited to the codebook index PMI1, PMI2, RI, CQI, reference signal received power RSRP, etc.

[0090] Step 2: The base station obtains uplink channel measurement information of the scheduled users UE1 and UE2, such as SRS reference signal measurement results. The uplink channel measurement information may include but is not limited to uplink channel (SRS H) projection energy, multipath DOA, channel angle spread, multipath delay, SRS SINR, and mobile speed.

[0091] Step 3: The base station side may input the acquired uplink channel measurement information and downlink channel measurement information of UE1 and UE2 into the PMI enhancement model respectively;

[0092] Step 4: The PMI enhancement model performs feature learning based on the input uplink channel measurement information and downlink channel measurement information, and outputs the enhanced downlink PMI codebook W 增强PMI 1 and W 增强PMI 2;

[0093] Step 5: The base station side can dynamically select UE1's W based on the timeliness condition and the principle of minimizing interference between paired UEs. 原始PMI 1 or W 增强PMI 1. UE2's W 原始PMI 2 or W 增强PMI 2. Make empty allocation pair judgment.

[0094] In this example, if the empty allocation pair condition is met, UE1 and UE2 can be grouped into an empty group, and the corresponding MU weight processing process can be performed when scheduling the downlink PDSCH; if the empty allocation pair condition is not met, UE1 and UE2 can be frequency-division scheduled separately to improve the multi-user downlink experience rate.

[0095] The above-mentioned Example 1 for the SU-MIMO system and Example 2 for the MU-MIMO system can optimize the downlink PMI forming gain of users in the cell, thereby achieving the purpose of increasing cell throughput and improving user downlink experience.

[0096] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus the necessary general hardware platform, and of course it can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for enabling a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods described in each embodiment of the present application.

[0097] An embodiment of the present application further provides a computer-readable storage medium, in which a computer program is stored. The computer program is configured to execute the steps of any of the above method embodiments when run.

[0098] In an exemplary embodiment, the computer-readable storage medium may include, but is not limited to, various media that can store computer programs, such as a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk, or an optical disk.

[0099] An embodiment of the present application further provides an electronic device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0100] In an exemplary embodiment, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor, and the input / output device is connected to the processor.

[0101] An embodiment of the present application further provides a computer program product, comprising a computer program, wherein the computer program is configured by a processor to run the computer program to perform the steps of any one of the above method embodiments.

[0102] For specific examples in this embodiment, reference may be made to the examples described in the above embodiments and exemplary implementation modes, and this embodiment will not be described in detail here.

[0103] Obviously, those skilled in the art should understand that the modules or steps of the present application described above can be implemented using a general-purpose computing device, they can be concentrated on a single computing device, or distributed across a network composed of multiple computing devices, they can be implemented using program code executable by the computing device, and thus, they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be performed in a different order than herein, or they can be fabricated into separate integrated circuit modules, or multiple modules or steps can be fabricated into a single integrated circuit module for implementation. Thus, the present application is not limited to any specific combination of hardware and software.

[0104] The above description is merely an exemplary embodiment of the present application and is not intended to limit the present application. Those skilled in the art will readily appreciate that various modifications and variations are possible. Any modifications, equivalent substitutions, or improvements made within the principles of the present application shall be included within the scope of protection of the present application.

Claims

1. A method for constructing a downlink codebook, characterized in that: include: Obtaining uplink channel measurement information of a current user equipment UE and downlink channel measurement information of the current UE corresponding to a precoding matrix indication PMI codebook; Inputting the uplink channel measurement information and the downlink channel measurement information into a pre-trained PMI enhancement model to obtain a PMI enhancement weight output by the PMI enhancement model; The PMI codebook is constructed according to the PMI enhancement weight, wherein the constructed PMI codebook is used to schedule downlink channel data of the current UE.

2. The method according to claim 1, characterized in that The downlink channel measurement information of the UE includes an original PMI weight, and the method further includes: Parsing the channel state information CSI code stream reported by the UE according to the precoding matrix indication PMI codebook pre-configured for the UE; Constructing a PMI original weight of the PMI codebook according to the CSI code stream.

3. The method according to claim 2, characterized in that The current UE is a single UE, and the method further includes: According to the timeliness condition and the principle of maximizing spectrum efficiency, the PMI original weight or the PMI enhanced weight is selected from the PMI codebook, and single UE downlink channel data scheduling is performed according to the PMI original weight or the PMI enhanced weight.

4. The method according to claim 2, characterized in that The current UE is a multiple UE, and the method further includes: In response to the multiple UEs satisfying the UE empty allocation pair condition, selecting the PMI original weight or the PMI enhanced weight of the paired UE from the PMI codebook for user downlink channel data scheduling according to the timeliness condition and the principle of minimizing interference between paired UEs; or, In response to the multiple UEs not satisfying the UE empty allocation pair condition, according to the timeliness condition and the principle of maximizing spectrum efficiency, the PMI original weights or the PMI enhanced weights corresponding to the multiple UEs are respectively selected from the PMI codebook, and multi-UE downlink channel data scheduling is performed according to the PMI original weights or the PMI enhanced weights.

5. The method according to claim 1, wherein The PMI enhancement model is obtained as follows: Obtain the training data set for the PMI enhancement model to be trained; Setting the constraints and solution objectives of the PMI enhancement model to be trained; The PMI enhancement model to be trained is trained according to the training data set, the constraint conditions and the solution goal to obtain the PMI enhancement model.

6. The method according to claim 5, characterized in that Before obtaining a training data set for the PMI enhancement model to be trained, the method further includes: Obtain historical UE downlink channel measurement information, historical UE uplink channel measurement information and target PMI weight; A training data set for the to-be-trained PMI enhancement model is constructed according to the historical UE downlink channel measurement information, the historical UE uplink channel measurement information, and the target PMI weight.

7. The method according to claim 6, characterized in that The historical UE downlink channel measurement information includes a PMI reported by the UE, and constructing a training data set for the PMI enhancement model to be trained according to the historical UE downlink channel measurement information, the historical UE uplink channel measurement information, and the target PMI weight, including: Selecting, according to the time corresponding to the UE reporting the PMI, historical UE uplink channel measurement information whose measurement time is close to the time corresponding to the UE reporting the PMI; The historical UE downlink channel measurement information, the historical UE uplink channel measurement information whose measurement time is close to the time corresponding to the UE reporting the PMI, and the target PMI weight are combined to construct a training data set for the PMI enhancement model to be trained.

8. The method according to any one of claims 5 to 7, characterized in that: The downlink channel includes multiple data layers, wherein the constraints include: orthogonalization of PMI enhancement weights of the multiple data layers output by the PMI enhancement model; and the solution objectives include: maximizing the correlation between the PMI enhancement weights output by the PMI enhancement model and the target PMI weights.

9. The method according to claim 1, characterized in that The UE downlink channel measurement information also includes at least one of the following: PMI, rank index RI, channel quality indicator CQI, and reference signal received power RSRP reported by the UE.

10. The method according to claim 1, characterized in that The UE uplink channel measurement information includes at least one of the following: uplink channel projection energy calculated by the user uplink sounding reference signal SRS, multipath arrival angle direction DOA, channel angle spread, multipath delay, sounding reference signal signal to interference plus noise ratio SRS SINR and UE moving speed.

11. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, wherein the computer program implements the steps of the method described in any one of claims 1 to 10 when executed by a processor.

12. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 10 are implemented.

13. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 10 are implemented.

Images