User side visual area map construction method for super-large scale MIMO (Multiple Input Multiple Output)

Abstract

The invention discloses a super-large-scale MIMO-oriented user side visual area map construction method, which comprises the following steps of: mapping a user side target area covered by an XL-MIMO antenna array into a map, and enabling each sampling position to correspond to a pixel point in the map; collecting user positions from a target area, sending uplink pilot signals, and measuring the receiving intensity of each pilot signal at a base station side; judging an effective communication link according to the receiving strength of each pilot signal, and taking an antenna unit corresponding to the effective communication link as a VR (Virtual Reality) of a sampling position; interpolating VR parameters of other positions based on the VR judgment result of the sampling position; and constructing a VR domain channel map corresponding to the user side coverage area based on VR parameters obtained by sampling measurement and interpolation prediction. According to the invention, an efficient low-complexity transmission scheme support can be provided for a super-large-scale MIMO system.

Description

Technical Field

[0001] This invention belongs to the field of communication technology, and in particular relates to a method for constructing user-side visible area maps for ultra-large-scale MIMO. Background Technology

[0002] Sixth-generation (6G) mobile communication systems aim to significantly improve the utilization of spatial resources, and ultra-large-scale multiple-input multiple-output (XL-MIMO), as a core technology of the physical layer, plays a crucial role in enhancing the system's spatial multiplexing capability. However, this technology faces challenges in practical applications due to the non-stationary propagation characteristics of electromagnetic waves in space. Specifically, XL-MIMO systems exhibit near-field visible area (VR) channel characteristics, where some antenna arrays are only "visible" to specific users. These antenna elements that can establish effective communication links with users are called the user's VR. In-depth research has revealed that fully utilizing this VR characteristic can not only achieve orthogonal transmission among multiple users but also effectively reduce the dimensionality of the channel matrix, thus providing new ideas for the optimized design of XL-MIMO systems. Therefore, the VR characteristic is of great significance in improving system performance. To leverage the role of VR, it is essential to accurately obtain the VR information of each user. However, existing methods require sending pilot signals and performing VR measurements individually for each user. This approach leads to enormous consumption of pilot resources, placing an unbearable burden on the system. Summary of the Invention

[0003] Purpose of the Invention: The purpose of this invention is to provide a method for constructing a user-side visible area map for ultra-large-scale MIMO. The method involves sending uplink pilot signals from a small number of locations sampled on the user side, receiving and measuring these signals via an ultra-large-scale MIMO antenna array on the base station side, and then determining the user VR parameters for each sampled location. Based on these parameters, the user VR parameters for the remaining locations are interpolated and predicted, thereby constructing a user-side VR domain channel map to guide the design of low-complexity transmission for ultra-large-scale MIMO.

[0004] Technical solution: The present invention provides a method for constructing user-side visible area maps for ultra-large-scale MIMO, comprising the following steps:

[0005] Step 1: Map the user-side target area covered by the XL-MIMO antenna array to a map, and assign each sampling location to a pixel on the map;

[0006] Step 2: Collect uplink pilot signals from the user location in the target area and measure the received strength of each pilot signal at the base station.

[0007] Step 3: Determine the valid communication link based on the received strength of each pilot signal, and use the antenna element corresponding to the valid communication link as the visible area VR of the sampling position;

[0008] Step 4: Interpolate the VR parameters of the remaining locations based on the VR decision results of the visible area at the sampling location;

[0009] Step 5: Based on the VR parameters obtained from sampling measurements and interpolation predictions, construct a VR domain channel map corresponding to the user-side coverage area.

[0010] Furthermore, step 1 specifically involves: covering the target area with the ultra-large-scale MIMO antenna array. Divided into equal intervals For each grid point, construct a planar channel map with the same number of pixels. And the number of rows and columns of the pixels is the same as the number of rows and columns of the grid points; the set of user locations mapped to all grid points is denoted as Each element in this set corresponds to a pixel in the channel map representing the user's location, and the attribute value of this pixel is the VR parameter of the corresponding grid point; target area. With channel map The mapping relationship between them is as follows:

[0011]

[0012] in, and These represent the number of rows and columns of grid points, respectively.

[0013] Furthermore, step 2 specifically involves: selecting a subset of user locations within the target coverage area and transmitting uplink probe pilot signals to the base station. Since signals transmitted by users in similar spatial locations will experience similar channel propagation environments, there is significant spatial continuity between the received signals. To reduce pilot signal transmission overhead, the user side only samples a subset of user locations. Send pilot signal; refine sampling by adjusting all sampling positions according to the scaling factor. It is divided into two parts: detection sampling and refinement sampling.

[0014] Furthermore, step 3 specifically involves: the base station-side ultra-large-scale MIMO antenna receiving and parsing the pilot signals sent by the sampling users, in order to sample the users' signals. Taking reception and decision-making as an example, the received signal strength measured by each antenna element is compared with a pre-set threshold value. The comparison determines whether each antenna element can establish a valid communication link with the sampling user; the VR identifier of the antenna element that can establish a valid communication link is set to one, and the others are set to zero; if the antenna element is known... Sampled users The received signal strength of the uplink pilot is The VR recognition decision expression is as follows:

[0015] , ,

[0016] in, Indicates the number of antenna elements. The decision threshold value representing the received signal is used to obtain the sampling user. VR parameters corresponding to the location:

[0017]

[0018] Due to vectors Each component stores a value of type Boolean, using Bit binary number b n ∈ [ 0 , 2 M − 1 ] Representation; scalar Each data point corresponds to dimensional vector One component, taking into account It is the result of the quantification of the judgment, therefore The values ​​are discrete;

[0019] To simplify antenna dimensions, the antenna array is divided into multiple non-overlapping antenna subarrays. The calculations for each subarray... All antenna elements Average value of received signal strength The expression used to determine the VR identifier corresponding to the subarray is as follows:

[0020] , ,

[0021] in, Indicates the number of antenna subarrays; , Subarray The number of antenna elements, g na Antenna element Receive from user The signal strength;

[0022] The sampled user VR parameters obtained by the base station decision provide data support for the subsequent VR spatial interpolation of users in other locations, and are used to assist the base station in measuring or inferring the VR parameters of all user locations within the coverage area;

[0023] The sampling dataset was constructed by combining the location of each sampled user with their VR parameters:

[0024]

[0025] in, and They represent the sampling users respectively. Location and its VR parameters, D s This represents the set of user locations sampled and measured.

[0026] Furthermore, step 4 specifically involves: utilizing the known spatial relationship between the sampled user's location and other locations, and combining this with the VR parameters obtained from the sampled location decision to perform spatial interpolation, thereby predicting the VR parameters for the remaining locations. Specifically, after obtaining the VR parameters for the sampled locations, the VR parameter dataset obtained from the sampled measurements is used as the basis for prediction. The VR parameters for the remaining locations are predicted using spatial interpolation techniques.

[0027]

[0028] in, and These represent the users to be tested. Location and its VR parameter estimates, This represents the set of locations of the remaining users to be tested. Represents the interpolation function;

[0029] When sampling dataset When fixed, the interpolation prediction effect and the interpolation function The only relevant factor is the VR parameters of users within the coverage area, which are spatially discrete. Therefore, the classic nearest neighbor interpolation method is used to predict the VR parameters of other users' locations, and the target user's VR parameters are calculated using nearest neighbor interpolation. The formula for the VR parameter is as follows:

[0030]

[0031] in, Indicates to users The nearest sampling user, Indicates user The VR parameters obtained from the measurements.

[0032] Furthermore, step 5 specifically involves: processing the VR parameters obtained from the location measurement decisions or interpolation predictions for each user. and Combined, we obtain the VR parameters corresponding to all user locations. This data is used as the value of the corresponding pixel in the map, i.e., the VR channel feature. By storing the VR parameters of all pixels, a user-side VR domain binary channel map is constructed.

[0033] M v = [ v 1 , 1 v 1 , 2 ⋯ v 1 , N c v 2 , 1 v 2 , 2 ⋯ v 2 , N c ⋮ ⋮ ⋱ ⋮ v N r , 1 v N r , 2 ⋯ v N r , N c ] ,

[0034] in, Indicates the first [item] in the constructed VR domain map Line number The VR parameters stored in the column pixels are the VR information corresponding to the user's position. and and The correspondence is ;

[0035] Since the channel environment for signal propagation is quasi-static, the VR domain channel map is dynamically updated periodically in practical applications. Within the user-side coverage area, uplink pilot signals are first periodically sampled from some user locations to obtain newly added sampled datasets.

[0036] ,

[0037] in, This represents the latest set of user locations; then it is combined with the original sampling measurement data. Perform spatial interpolation:

[0038] ,

[0039] in, This represents the interpolation function that combines the latest and historical sampling data, with the latest sampling data having a greater influence on the interpolation result. Finally, the VR decision is updated, and the VR domain channel map is dynamically updated based on the decision result.

[0040] M ˜ v = [ v ˜ 1 , 1 v ˜ 1 , 2 ⋯ v ˜ 1 , N c v ˜ 2 , 1 v ˜ 2 , 2 ⋯ v ˜ 2 , N c ⋮ ⋮ ⋱ ⋮ v ˜ N r , 1 v ˜ N r , 2 ⋯ v ˜ N r , N c ] ,

[0041] in, This indicates the updated VR judgment result.

[0042] The present invention also discloses a computer device, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the method of the present invention.

[0043] The present invention also discloses a computer-readable storage medium having a computer program / instructions stored thereon, which, when executed by a processor, implements the steps of the method of the present invention.

[0044] The present invention also discloses a computer program product, including a computer program / instructions that, when executed by a processor, implement the steps of the method of the present invention.

[0045] Beneficial effects: Compared with the prior art, the present invention has the following significant advantages:

[0046] 1. This invention proposes a method for constructing a user-side VR domain channel map for ultra-large-scale MIMO antenna array coverage. This method utilizes a base station-side ultra-large-scale MIMO antenna array to cover a target area. First, a small number of key locations are selected for sampling, and their received signal strength is measured to determine VR identifiers and generate VR parameters (used to characterize VR channel features). Then, utilizing the spatial continuity of channel features and combining spatial relationship interpolation, the VR parameters of the remaining locations are predicted, ultimately forming a complete user-side VR domain channel map for ultra-large-scale MIMO antenna array coverage. This method can quickly acquire all VR information within the target coverage area under low complexity conditions, significantly reducing pilot overhead and storage resource requirements. Using the constructed channel map, the base station can query the VR parameters of target users within its coverage area in real time and optimize the selection of effective transmission antenna elements for target users accordingly, thereby providing efficient low-complexity transmission scheme support for ultra-large-scale MIMO systems.

[0047] 2. Analysis of the spatial location characteristics of signal receiving points within the coverage area of ​​a large-scale MIMO antenna array reveals that users in similar spatial regions exhibit similar channel transmission characteristics. Based on this characteristic, a phased approach can be adopted in the channel estimation process: by deploying uplink pilot signals at some core user locations to obtain measurement results of the corresponding VR channel characteristics, and then using interpolation algorithms to predict the VR characteristics at the remaining user locations, the high overhead of needing to send pilot signals for each possible location can be avoided.

[0048] 3. Considering that although the radio electromagnetic propagation environment exhibits dynamic characteristics over time, it can be approximated as a quasi-static state within a relatively short time period. This means that by storing the current VR information of different users in the same area using location parameters as indexes, a VR domain channel map can be constructed, providing a basis for subsequent rapid retrieval and acquisition of the VR characteristics of target users. This approach not only avoids the need for repeated measurements of VR channel characteristics but also further optimizes the utilization efficiency of pilot resources.

[0049] 4. User-side VR domain channel maps have significant advantages in achieving low-overhead acquisition of VR information. To fully utilize this technology to guide the low-complexity transmission design of ultra-large-scale MIMO systems, the primary task is to research and design efficient VR domain channel map construction methods to ensure the feasibility and effectiveness of its practical application.

[0050] 5. Low map construction overhead: This invention measures VR parameters by sampling only a small number of user locations within the coverage area, and predicts VR parameters for the remaining user locations using spatial interpolation techniques. Therefore, the overall pilot measurement overhead when constructing the channel map is relatively low. High VR prediction accuracy: On the one hand, a small-user sampling method combining detection and refinement is used to increase the overall amount of prior information; on the other hand, the nearest neighbor spatial interpolation method is used to rationally utilize prior information to improve VR prediction accuracy. Low data storage overhead: VR parameters are stored using antenna subarrays as basic storage units, significantly reducing the storage space occupied by the VR domain channel map while retaining effective VR information. Dynamic map updates are supported: The measurement results of pilot signals transmitted from sampled user locations can be updated periodically, supporting dynamic updates of the VR domain channel map, thereby adapting to the slow changes in the channel environment. Attached Figure Description

[0051] Figure 1 This is a schematic diagram of the overall concept of the present invention.

[0052] Figure 2 This is a flowchart illustrating the key technologies for constructing a user-side VR domain channel map for ultra-large-scale MIMO antenna array coverage, as described in this invention.

[0053] Figure 3 This diagram illustrates the sampling and measurement of pilot signals from a small number of user locations within the coverage area of ​​this ultra-large-scale MIMO antenna array.

[0054] Figure 4 This is a schematic diagram illustrating the compression of VR data storage space by dividing the antenna subarray.

[0055] Figure 5 Performance curves of VR domain channel map construction based on different sampling methods as the number of sampling points increases.

[0056] Figure 6 Performance curves of VR domain channel map construction based on different VR parameter storage methods as the number of sampling points increases.

[0057] Figure 7 This is an example diagram of a user-side VR domain channel map for a very large-scale MIMO antenna array. Detailed Implementation

[0058] The technical solution of the present invention will be further described below with reference to the accompanying drawings.

[0059] This invention proposes a method for constructing a user-side VR domain channel map for UML antenna array coverage. This method utilizes a base station-side UML antenna array to cover a target area. First, a small number of key locations are sampled to measure their received signal strength, determining VR identifiers and generating VR parameters (used to characterize VR channel features). Then, leveraging the spatial continuity of channel features and combining spatial relationship interpolation, the VR parameters for the remaining locations are predicted, ultimately forming a complete user-side VR domain channel map for UML antenna array coverage. This method enables rapid acquisition of all VR information within the target coverage area under low complexity conditions, significantly reducing pilot overhead and storage resource requirements. Using the constructed channel map, the base station can query the VR parameters of target users within its coverage area in real time and optimize the selection of effective transmission antenna elements for target users, thereby providing efficient, low-complexity transmission scheme support for UML systems.

[0060] Since the VR channel characteristics of users within the coverage area of ​​a very large-scale MIMO antenna array exhibit spatial continuity, this invention proposes a method for constructing a VR domain channel map based on this characteristic. Specifically, firstly, pilot signals are transmitted from a small number of user locations within the antenna array coverage area and received at the base station, and their signal strength is measured. Next, the VR identifier corresponding to each antenna element is determined. Subsequently, spatial interpolation technology is used to predict the VR identifiers for the remaining user locations. Finally, all user VR identifiers obtained from sampling decisions and interpolation predictions are integrated to construct a complete user-side VR domain channel map. The core of this method lies in utilizing the spatial correlation of VR parameters between different user locations to achieve low-overhead VR channel feature acquisition and storage while ensuring accuracy. This requires that this method fully utilize and design an interpolation prediction mechanism based on spatial continuity to ensure that VR distribution information within the entire target coverage area can be efficiently inferred using only a small number of sampling points.

[0061] like Figure 1 As shown, the overall approach to constructing a user-side VR domain channel map for ultra-large-scale MIMO antenna array coverage includes key steps such as target area map mapping, pilot transmission at sampling locations, pilot measurement and user VR decision, VR interpolation at other locations, and user-side VR domain map construction.

[0062] Key technology implementation methods such as Figure 2As shown, firstly, the user-side target area covered by the ultra-large-scale MIMO antenna array is mapped as a channel map, with each possible user location corresponding to a pixel in the channel map. Secondly, a small number of user locations within the target area are sampled and uplink pilot signals are transmitted, and the received strength of each pilot signal is measured at the base station. Next, based on the measured received signal strength, antenna array elements that can establish effective communication links with each sampling location are identified, and the antenna elements corresponding to the effective communication links are determined as the user VRs at the sampling locations. Then, the VR parameters of the remaining locations are predicted by interpolation based on the VR determination results (i.e., VR parameters) of the sampling locations. Finally, the VR parameters obtained from the sampling measurements and the interpolation prediction are combined to construct the user-side VR domain channel map corresponding to the coverage area.

[0063] The specific process is as follows:

[0064] 1. Target area map mapping

[0065] The target area covered by the ultra-large-scale MIMO antenna array Divided into equal intervals For each grid point, construct a planar channel map with the same number of pixels. Furthermore, the number of rows and columns of the pixels is the same as the number of rows and columns of the grid points. Let the set of user locations mapped to all grid points be denoted as... Each element (user location) in this set corresponds to a pixel in the channel map, and the pixel's attribute value is the VR parameter of the corresponding grid point. Target area With channel map The mapping relationship between them is as follows:

[0066]

[0067] in, and These represent the number of rows and columns of grid points, respectively.

[0068] 2. Pilot transmission at sampling location

[0069] Within the target coverage area, a small number of user locations are selected to transmit uplink probe pilot signals to the base station. Since signals transmitted by users in spatially close proximity are highly likely to experience similar channel propagation environments, there is significant spatial continuity between the received signals. To reduce pilot signal transmission overhead, the user side can sample only a small number of user locations. Send pilot signal.

[0070] The accuracy of the VR domain channel map construction on the user side is related to the density of sampled user locations. Generally, the higher the sampling density, the higher the accuracy of the constructed VR domain map. However, a higher sampling density also leads to greater pilot measurement overhead. To balance map construction accuracy and pilot measurement overhead, the sampling strategy needs to be optimized to maximize map accuracy at a given sampling density. Under the constraint of a limited number of sampling points, a sampling strategy combining probing and refinement can improve sampling efficiency and increase map construction accuracy without increasing pilot measurement overhead. Probing and refinement sampling scales all sampling locations according to a scaling factor. The sampling process is divided into two parts: probe sampling and refinement sampling. On the one hand, in order to fully explore the distribution characteristics of the received signal strength of the antenna array, probe sampling can use uniform spatial sampling to quickly obtain the overall VR channel feature profile. On the other hand, in order to accurately depict the edge details of the received signal strength of the antenna array, refinement sampling can use non-uniform spatial sampling to continuously enrich the feature details of each VR channel, thereby mining as much VR channel information as possible and providing more prior knowledge for subsequent spatial interpolation.

[0071] 3. Pilot measurement and user VR decision

[0072] The base station-side ultra-large-scale MIMO antenna receives and analyzes the pilot signals sent by the sampling users in order to sample the users. Taking reception and decision-making as an example, the received signal strength measured by each antenna element is compared with a pre-set threshold value. The comparison determines whether each antenna element can establish a valid communication link with the sampling user. The VR flag of the antenna elements that can establish a valid communication link is set to one, and the rest are set to zero. If the antenna elements are known... Sampled users The received signal strength of the uplink pilot is The VR recognition decision expression is as follows:

[0073] , ,

[0074] in, Indicates the number of antenna elements. This represents the decision threshold value for the received signal. Based on this, the sampling user can be obtained. VR parameters corresponding to the location:

[0075]

[0076] Due to vectors Each component stores a Boolean value (either 1 or 0), therefore it can be used... Bit binary number b n ∈ [ 0 , 2 M − 1 ] Representation. Scalar Each data point corresponds to dimensional vector One component. Furthermore, considering... It is the result of the quantification of the judgment, therefore The values ​​are discrete.

[0077] Due to the large size of the antenna array, Bit binary number This indicates that VR parameters occupy a large amount of storage space, which can easily lead to data storage disasters and is not conducive to the efficient construction of VR domain channel maps. Therefore, it is necessary to simplify the antenna dimensions. One feasible method is to divide the antenna array into multiple non-overlapping antenna subarrays on an even scale, and then calculate the parameters for each subarray. All antenna elements Average value of received signal strength The expression used to determine the VR identifier corresponding to the subarray is as follows:

[0078] , ,

[0079] in, Indicates the number of antenna subarrays; , Subarray The number of antenna elements.

[0080] The VR parameters of the sampled users obtained by the base station decision can provide data support for the VR spatial interpolation of users in other locations, and can be used to assist the base station in measuring or inferring the VR parameters of all user locations within the coverage area.

[0081] The sampling dataset can be constructed by combining the location of each sampled user with their VR parameters:

[0082]

[0083] in, and They represent the sampling users respectively. The location and its VR parameters.

[0084] 4. VR interpolation at other locations

[0085] Because the VR channel characteristics within the coverage area of ​​a very large-scale MIMO antenna array exhibit spatial continuity, the VR channel characteristics corresponding to users in spatially close locations show a stable variation trend. By utilizing the known spatial relationship between the sampled user's location and other locations, and combining this with the VR parameters obtained from the sampling location decision, spatial interpolation can be performed to predict the VR parameters for the remaining locations. Specifically, after obtaining the VR parameters for the sampling locations, the VR parameter dataset obtained from the sampling measurements is used... Spatial interpolation techniques can be used to predict the VR parameters at other locations:

[0086]

[0087] in, and These represent the users to be tested. Location and its VR parameter estimates, This represents the set of locations of the remaining users to be tested. This represents the interpolation function.

[0088] When sampling dataset When fixed, the interpolation prediction effect and the interpolation function Unique correlation. Since the VR parameters of users within the coverage area are spatially discrete, it is impossible to predict them jointly using measurement data from multiple sampled users. Furthermore, the correlation of VR channel characteristics between a sampled user and other users is inversely proportional to the distance between them. Therefore, the classic nearest neighbor (NN) interpolation method can be used to predict the VR parameters of other users' locations. The target user is calculated using nearest neighbor interpolation. The formula for the VR parameter is as follows:

[0089]

[0090] in, Indicates to users The nearest sampling user, Indicates user The VR parameters obtained from the measurements.

[0091] 5. User-side VR domain map construction

[0092] VR parameters obtained from the measurement decisions or interpolation predictions of each user's location and Combined, we obtain the VR parameters corresponding to all user locations. This information is then used as the value of the corresponding pixel in the map (i.e., VR channel features). By storing the VR parameters of all pixels, a user-side VR domain binary channel map can be constructed, i.e.

[0093] M v = [ v 1 , 1 v 1 , 2 ⋯ v 1 , N c v 2 , 1 v 2 , 2 ⋯ v 2 , N c ⋮ ⋮ ⋱ ⋮ v N r , 1 v N r , 2 ⋯ v N r , N c ] ,

[0094] in, Indicates the first [item] in the constructed VR domain map Line number The column of pixels stores VR parameters, which are the VR information corresponding to the user's position. Parameters and and The correspondence is The VR domain channel map reflects the ability of each user location to establish an effective communication link with the base station's ultra-large-scale MIMO antenna array. This map allows for the rapid querying and acquisition of target user VR information, enabling the design of low-complexity transmission schemes.

[0095] Because the channel environment for signal propagation is quasi-static, meaning it changes slowly, it is necessary to periodically update the VR domain channel map dynamically in practical applications. Specifically, within the user-side coverage area, a small number of user locations are periodically sampled to transmit uplink pilot signals, thereby obtaining a newly added sampling dataset:

[0096] ,

[0097] in, This represents the latest set of user locations; then it is combined with the original sampling measurement data. Perform spatial interpolation

[0098] ,

[0099] in, This represents the interpolation function that combines the latest and historical sampled data. Typically, the latest sampled data has a greater influence on the interpolation result. Finally, the VR decision is updated, and the VR domain channel map is dynamically updated based on the decision result.

[0100] M ˜ v = [ v ˜ 1 , 1 v ˜ 1 , 2 ⋯ v ˜ 1 , N c v ˜ 2 , 1 v ˜ 2 , 2 ⋯ v ˜ 2 , N c ⋮ ⋮ ⋱ ⋮ v ˜ N r , 1 v ˜ N r , 2 ⋯ v ˜ N r , N c ] ,

[0101] in, This indicates the updated VR judgment result.

[0102] Example 1

[0103] To further illustrate key processes such as target area map mapping, pilot transmission for sampled user locations, base station antenna reception measurement, and user VR decision-making, this case study constructs an ultra-large-scale MIMO near-field communication coverage scenario, such as... Figure 3As shown in the figure, the left side represents the base station's ultra-large-scale MIMO antenna array, and the right side represents the target area within the coverage of this antenna array. The target area periodically transmits uplink pilot signals to the base station using sampled users. To reduce pilot estimation overhead, the user side only samples a small number of user locations within the coverage area (shown as blue squares in the figure) and transmits uplink pilot signals. The base station's antenna array then receives and measures the pilot signals and determines the VR parameters corresponding to these users. The remaining numerous user locations (shown as gray squares in the figure) undergo spatial interpolation based on the sampling decision results and their own locations to obtain the VR parameters of all grid points within the entire coverage area.

[0104] Example 2

[0105] To reduce the storage space occupied by user VR data in the VR domain channel map, this case study presents a design approach to compress VR data storage space by dividing the antenna subarray, as illustrated in the diagram below. Figure 4 As shown in the diagram, assuming the VMIMO antenna array contains M=256 antenna elements, the left sub-diagram shows an VMIMO antenna array that uses the original antenna elements as VR identification units. The VR measurement identification results (1 or 0) of users within its coverage area are stored using antenna elements as basic units, with each antenna element corresponding to 1 bit of storage space. Due to the large scale of the antenna array, storing the VR identification results of the entire antenna array for a single user requires 256 bits of storage space, resulting in significant overhead. In contrast, the right sub-diagram shows a VMIMO antenna array that uses sub-arrays as VR identification units (antenna elements of different colors belong to different sub-arrays). The VR measurement identification results (1 or 0) of users within its coverage area are stored using sub-arrays as units, with each sub-array corresponding to 1 bit of storage space. Taking an antenna array divided into L=16 sub-arrays as an example, storing the VR identification results of the entire antenna array for a single user only requires 16 bits of storage space, significantly reducing overhead. The VR identification of each sub-array requires integrating all its antenna elements, first by averaging the received signal strengths of these antenna elements. Then, with the VR decision threshold In comparison, subarrays with a mean greater than the threshold value can be identified as having a VR identifier of 1.

[0106] Example 3

[0107] To verify the performance of the proposed probe refinement sampling method in the VR domain channel map construction process, this case study simulates the VR prediction accuracy under three methods: random sampling, uniform sampling, and probe refinement sampling. The simulation process is completed in five batches. The first batch randomly samples 200 grid points to measure their VR parameters, and each subsequent batch adds another 200 sampling points to measure their VR parameters. The channel map stores VR parameters based on antenna elements, with the probe refinement sampling scaling factor set to 0.5. Using random sampling as the baseline, the performance curves of the accuracy of predicting the VR parameters of other users using the sample points selected by the three spatial sampling schemes as the number of samples increases are shown below. Figure 5 As shown in the figure, the proposed refined sampling scheme outperforms other methods with different numbers of sampling points. Its sampling results better support the nearest neighbor interpolation algorithm for accurate prediction of VR parameters, and this performance advantage is particularly pronounced when the number of sampling points is limited. The above analysis demonstrates that the simulation results validate the effectiveness of the proposed refined sampling method.

[0108] Example 4

[0109] To verify the performance of the proposed subarray-based VR parameter storage method in VR domain channel map construction, this case study simulates both antenna element-based and subarray-based VR parameter storage methods. The simulation process was also completed in five batches. The first batch randomly sampled 200 grid points to measure their VR parameters, and each subsequent batch added 200 sampling points to measure their VR parameters. The more expensive antenna element-based VR parameter storage method was used as the benchmark comparison. The performance curves of the accuracy of the VR domain channel map construction using the two VR parameter storage methods as the number of samples increases are shown below. Figure 6 As shown in the figure, under different numbers of sampling points, the proposed subarray-based VR parameter storage method performs close to the antenna element-based VR parameter storage method, maintaining an overall VR prediction accuracy of over 90%. Its prediction results ensure the effectiveness of VR information with low storage overhead. The above analysis demonstrates that the current simulation results validate the effectiveness of the proposed subarray-based VR parameter storage method.

[0110] Example 5

[0111] To further demonstrate the effectiveness of user-side VR domain channel map construction, this case study presents a possible result for constructing a user-side VR domain channel map for coverage of a very large-scale MIMO antenna array, such as... Figure 7As shown in the figure, this is a VR domain channel map storing VR parameters. Each small square in the figure represents a pixel in the VR domain channel map, and the hexadecimal value of the color block of the square represents the VR parameter value of that pixel, that is, the status of the target user establishing a valid communication link with the antenna array; different colors correspond to different VR labels. Taking the VR parameter 0x7FFE corresponding to the blue area in the middle as an example, the corresponding binary value is "011111111111110", where the number of VR identifiers "1" is 14, indicating that this area can establish a valid communication link with the 14 subarrays in the VMIMO antenna array. Using this map, the VR channel characteristics of the antenna array can be quickly queried and inferred, thereby guiding the low-complexity transmission design of VMIMO and providing users with better communication service quality.

Claims

1. A method for constructing user-side visible area maps for ultra-large-scale MIMO, characterized in that, Includes the following steps: Step 1: Map the user-side target area covered by the XL-MIMO antenna array to a map, and assign each sampling location to a pixel on the map; Step 2: Collect uplink pilot signals from the user location in the target area and measure the received strength of each pilot signal at the base station. Step 3: Determine the valid communication link based on the received strength of each pilot signal, and use the antenna element corresponding to the valid communication link as the visible area VR of the sampling position; Step 4: Interpolate the VR parameters of the remaining locations based on the VR decision results of the visible area at the sampling location; Step 5: Based on the VR parameters obtained from sampling measurements and interpolation predictions, construct a VR domain channel map corresponding to the user-side coverage area.

2. The method for constructing a user-side visible area map for ultra-large-scale MIMO according to claim 1, characterized in that, Step 1 specifically involves: covering the target area with the ultra-large-scale MIMO antenna array. Divided into equal intervals For each grid point, construct a planar channel map with the same number of pixels. And the number of rows and columns of the pixels is the same as the number of rows and columns of the grid points; the set of user locations mapped to all grid points is denoted as Each element in this set corresponds to a pixel in the channel map representing the user's location, and the attribute value of this pixel is the VR parameter of the corresponding grid point; target area. With channel map The mapping relationship between them is as follows: ； in, and These represent the number of rows and columns of grid points, respectively.

3. The method for constructing a user-side visible area map for ultra-large-scale MIMO according to claim 1, characterized in that, Step 2 specifically involves selecting a subset of user locations within the target coverage area and transmitting uplink probe pilot signals to the base station. Since signals transmitted by users in similar spatial locations will experience similar channel propagation environments, there is significant spatial continuity between the received signals. To reduce pilot signal transmission overhead, the user side only samples a subset of user locations. Send pilot signal; refine sampling by adjusting all sampling positions according to the scaling factor. It is divided into two parts: detection sampling and refinement sampling.

4. The method for constructing a user-side visible area map for ultra-large-scale MIMO according to claim 2, characterized in that, Step 3 specifically involves: the base station-side ultra-large-scale MIMO antenna receiving and parsing the pilot signals sent by the sampling users, in order to sample the users' signals. Taking reception and decision-making as an example, the received signal strength measured by each antenna element is compared with a pre-set threshold value. The comparison determines whether each antenna element can establish a valid communication link with the sampling user; the VR identifier of the antenna element that can establish a valid communication link is set to one, and the others are set to zero; if the antenna element is known... Sampled users The received signal strength of the uplink pilot is The VR recognition decision expression is as follows: ， ， ； in, Indicates the number of antenna elements. The decision threshold value representing the received signal is used to obtain the sampling user. VR parameters corresponding to the location: ； Due to vectors Each component stores a value of type Boolean, using Bit binary number Representation; scalar Each data point corresponds to dimensional vector One component, taking into account It is the result of the quantification of the judgment, therefore The values ​​are discrete; To simplify antenna dimensions, the antenna array is divided into multiple non-overlapping antenna subarrays. The calculations for each subarray... All antenna elements Average value of received signal strength The expression used to determine the VR identifier corresponding to the subarray is as follows: ， ， ； in, Indicates the number of antenna subarrays; , Subarray The number of antenna elements, g na Antenna element Receive from user The signal strength; The sampled user VR parameters obtained by the base station decision provide data support for the subsequent VR spatial interpolation of users in other locations, and are used to assist the base station in measuring or inferring the VR parameters of all user locations within the coverage area; The sampling dataset was constructed by combining the location of each sampled user with their VR parameters: ； in, and They represent the sampling users respectively. Location and its VR parameters, D s This represents the set of user locations sampled and measured.

5. The method for constructing a user-side visible area map for ultra-large-scale MIMO according to claim 4, characterized in that, Step 4 specifically involves: using the known spatial relationship between the sampled user's location and other locations, and combining this with the VR parameters obtained from the sampled location decision, performing spatial interpolation to predict the VR parameters for the remaining locations. Specifically, after obtaining the VR parameters for the sampled locations, the VR parameter dataset obtained from the sampled measurements is used... The VR parameters for the remaining locations are predicted using spatial interpolation techniques. ； in, and These represent the users to be tested. Location and its VR parameter estimates, This represents the set of locations of the remaining users to be tested. Represents the interpolation function; When sampling dataset When fixed, the interpolation prediction effect and the interpolation function The only relevant factor is the VR parameters of users within the coverage area, which are spatially discrete. Therefore, the classic nearest neighbor interpolation method is used to predict the VR parameters of other users' locations, and the target user's VR parameters are calculated using nearest neighbor interpolation. The formula for the VR parameter is as follows: ； in, Indicates to users The nearest sampling user, Indicates user The VR parameters obtained from the measurements.

6. The method for constructing a user-side visible area map for ultra-large-scale MIMO according to claim 5, characterized in that, Step 5 specifically involves: assembling the VR parameters obtained from the location measurement decisions or interpolation predictions for each user. and Combined, we obtain the VR parameters corresponding to all user locations. This data is used as the value of the corresponding pixel in the map, i.e., the VR channel feature. By storing the VR parameters of all pixels, a user-side VR domain binary channel map is constructed. ， in, Indicates the first [item] in the constructed VR domain map Line number The VR parameters stored in the column pixels are the VR information corresponding to the user's position. and and The correspondence is ; Since the channel environment for signal propagation is quasi-static, the VR domain channel map is dynamically updated periodically in practical applications. Within the user-side coverage area, uplink pilot signals are first periodically sampled from some user locations to obtain newly added sampled datasets. ， in, This represents the latest set of user locations; then it is combined with the original sampling measurement data. Perform spatial interpolation: ， in, This represents the interpolation function that combines the latest and historical sampling data, with the latest sampling data having a greater influence on the interpolation result. Finally, the VR decision is updated, and the VR domain channel map is dynamically updated based on the decision result. ， in, This indicates the updated VR judgment result.

7. A computer device comprising a memory, a processor, and a computer program stored in the memory, characterized in that, The processor executes the computer program to implement the steps of the method of claim 1.

8. A computer-readable storage medium having a computer program / instructions stored thereon, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the method of claim 1.

9. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the method of claim 1.

Images