Near-field beam training method and device for single-user distributed MIMO system

Abstract

The invention discloses a near-field beam training method and device for a single-user distributed MIMO system, and the method comprises the steps: carrying out the far-field layered beam training of a corresponding AP based on a far-field layered codebook of each AP in the distributed MIMO system, and obtaining a beam direction of each AP pointing to a user; screening the positions of the users according to the plurality of beam directions to obtain position information of the users relative to the distributed MIMO system; according to the method, beam training is carried out on a distributed MIMO system by using a far-field codebook to estimate a user angle, and then the position of a user is determined by using a geometrical relationship according to the user angle; through the method, only time resources need to be occupied in the user angle estimation process, and compared with an existing near-field beam training technology, the beam training overhead is remarkably reduced.

Description

Technical Field

[0001] The present invention belongs to the technical field of near-field beam training, and particularly relates to a near-field beam training method and device for a single-user distributed MIMO system. Background Art

[0002] In order to meet the requirements of 6G communication for ultra-high speed, ultra-low latency, ultra-high reliability, and large-scale connection, distributed multiple-input multiple-output (MIMO) has attracted wide attention. Compared with a centralized MIMO system, a distributed MIMO system has significant advantages in terms of anti-interference and coverage. By deploying a large number of access points (APs) at different geographical locations for signal processing and cooperation, it provides stronger anti-interference ability and wider signal coverage, thereby improving the reliability and performance of the communication system.

[0003] Currently, with the significant increase in the number of antennas and carrier frequency in 6G systems, the Rayleigh distance has significantly expanded, and the near-field characteristics have become more obvious, making traditional far-field beam training methods no longer applicable to near-field MIMO channels. In a centralized MIMO system, inspired by the near-field beam focusing phenomenon, a polar codebook is constructed to divide the near-field space into a two-dimensional grid of angles and distances, and the beam training overhead is the number of two-dimensional grids divided. Moreover, if the beam training method of a centralized MIMO system is directly adopted in a distributed MIMO system, it will result in a much larger training overhead. Summary of the Invention

[0004] The purpose of the present invention is to provide a near-field beam training method and device for a single-user distributed MIMO system to reduce the near-field beam training overhead of the distributed MIMO system.

[0005] The present invention adopts the following technical solutions: A near-field beam training method for a single-user distributed MIMO system includes the following steps:

[0006] Based on the far-field hierarchical codebooks of each AP in the distributed MIMO system, perform far-field hierarchical beam training on the corresponding AP respectively to obtain the beam direction of each AP pointing to the user;

[0007] According to several beam directions, screen the position of the user to obtain the position information of the user relative to the distributed MIMO system.

[0008] Further, after obtaining the position information of the user relative to the distributed MIMO system, it further includes:

[0009] Calculate the near-field optimal codeword corresponding to each AP based on the position information.

[0010] Further, screening the user's position according to a plurality of beam directions includes:

[0011] Screening the user's position according to the geometric relationship of a plurality of beam directions.

[0012] Further, screening the user's position according to the geometric relationship of a plurality of beam directions includes:

[0013] Determining a plurality of intersection points according to a plurality of beam directions;

[0014] Based on the position information of each intersection point in the service area of the distributed MIMO system, screening out the intersection points in the position dense area to form an intersection point set;

[0015] Calculating the average position information based on the position information of the intersection points in the intersection point set;

[0016] Taking the average position information as the position information of the user relative to the distributed MIMO system.

[0017] Further, before determining a plurality of intersection points according to a plurality of beam directions, it further includes:

[0018] Generating symmetric beam directions symmetric about the corresponding AP according to the beam directions;

[0019] Selecting a beam direction or a symmetric beam direction for each AP as the final beam direction of the AP.

[0020] Further, selecting a beam direction or a symmetric beam direction for each AP includes:

[0021] Generating 2 M beam direction combinations according to the beam direction or the symmetric beam direction of each AP;

[0022] Selecting the optimal beam direction combination according to the average distance between the intersection points formed by the beam directions in the beam direction combinations;

[0023] Selecting a beam direction or a symmetric beam direction for each AP according to the optimal beam direction combination.

[0024] Further, selecting the optimal beam direction combination according to the average distance between the intersection points formed by the beam directions in the beam direction combinations includes:

[0025] Selecting the beam direction combination corresponding to the minimum value in the average distance as the optimal beam direction combination.

[0026] Further, selecting the beam direction combination corresponding to the minimum value in the average distance as the optimal beam direction combination includes:

[0027] Selecting the minimum value in the average distance based on the KD-Tree.

[0028] Further, the intersection points in the position dense area are screened out by using a density-based clustering algorithm.

[0029] Another technical solution of the present invention: A near-field beam training device for a single-user distributed MIMO system, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the above method is implemented.

[0030] The beneficial effects of the present invention are as follows: The present invention performs beam training using a far-field codebook for a distributed MIMO system to estimate the user's angle, and then determines the user's position based on the geometric relationship according to the user's angle; through this method, only the time resource is occupied during the user angle estimation process, and the beam training overhead is reduced compared to the hierarchical near-field beam training in the prior art. Description of the Drawings

[0031] Figure 1 It is a schematic diagram of the downlink of a single-user distributed MIMO system in an embodiment of the present invention;

[0032] Figure 2 It is a schematic diagram of the far-field beam training process based on the angular domain in an embodiment of the present invention;

[0033] Figure 3 It is a flow chart of user position screening in an embodiment of the present invention;

[0034] Figure 4 It is a performance analysis diagram of the achievable rate of different beam training methods relative to the signal-to-noise ratio in an embodiment of the present invention. Detailed Embodiments

[0035] The present invention will be described in detail below with reference to the drawings and specific embodiments.

[0036] In the existing research on near-field beam training, it is mainly carried out under a centralized MIMO system. If the beam training method of the centralized MIMO system is directly adopted in the distributed MIMO system, it will lead to a more huge training overhead and the system advantages of the distribution of multiple APs in space are not fully utilized. Therefore, how to design a near-field beam training method in a single-user distributed MIMO system has become an urgent problem to be solved.

[0037] The present invention discloses a near-field beam training method for a single-user distributed MIMO system, including the following steps: performing far-field hierarchical beam training on each AP in the distributed MIMO system based on the far-field hierarchical codebook of each AP to obtain the beam direction of each AP pointing to the user; screening the position of the user according to several beam directions to obtain the position information of the user relative to the distributed MIMO system.

[0038] The present invention performs beam training using a far - field codebook for a distributed MIMO system to estimate the user's angle, and then determines the user's position based on the geometric relationship according to the user's angle. By this method, only the time resource is occupied in the process of estimating the user's angle, which reduces the beam training overhead compared with the hierarchical near - field beam training in the prior art.

[0039] After obtaining the position information of the user relative to the distributed MIMO system in the present invention, it further includes: calculating the near - field optimal codeword corresponding to each AP based on the position information. It can be seen that the present invention uses a far - field codebook to perform beam training on the APs of a distributed MIMO system in a near - field environment. When determining the user's position, the corresponding near - field optimal codeword is calculated according to the user's position information for data transmission, which can greatly reduce the beam training overhead in the beam training process.

[0040] In the present invention, a single cell is used as the service area of the distributed MIMO system. As Figure 1 shown, considering a single - cell distributed MIMO near - field downlink system, the system distributes M APs. In the case of circular deployment of APs, all APs are evenly distributed on a circular trajectory with a radius of R. Then, with the center of the service area as the origin, any direction as the x - axis, and the direction perpendicular to the x - axis as the y - axis, a service area coordinate system is established. In this coordinate system, the polar coordinates of the m - th AP are (R, φ m )=(R, 2π(m - 1) / M), and the rectangular coordinates are (Rcosφ m , Rsinφ m ), where m = 1, 2, …, M, and φ m represents the deployment angle of the m APs on the circular trajectory.

[0041] Each AP is deployed with a uniform linear array (ULA) composed of N antennas, and the total number of transmitting antennas of the distributed MIMO system is N t = MN, and the spacing between adjacent two antennas in each AP is equal to half - wavelength. In the embodiment of the present invention, it is set that each ULA array is arranged along the y - axis direction in the service area coordinate system. Then, the coordinates of the n - th antenna of the m - th AP are (Rcosφ m , Rsinφ m +δ n d), where is half - wavelength and λ is the wavelength. All APs serve a single - antenna user simultaneously. The signal received by the user is:

[0042]

[0043] where x represents the transmission symbol of the AP, represents the channel matrix between the user and the m - th AP. Denote the beam steering vector at the m-th AP, which is essentially a codeword selected from a predefined near-field codebook. Denote the additive white Gaussian noise with power σ 2 . Beam training is to measure the power of the received signal y and obtain the optimal codeword through a predefined codebook or corresponding calculations.

[0044] Since the XL-MIMO channel is generally dominated by several main paths, it is only necessary to search for the physical positions of the main paths through beam training, rather than obtaining explicit channel information. Therefore, the present invention will focus on the near-field LOS channel to find the optimal beam steering vector aligned with the main path.

[0045] If the polar coordinates of the user are (r, θ), where r represents the distance between the user and the origin of the service area coordinate system, and θ represents the angle between the line connecting the user and the origin and the x-axis, then the relative coordinates of the user with respect to the m-th AP are (x m , y m ) = (r cosθ - R cosφ m , r sinθ - R sinφ m ). Then the polar coordinates of the user with respect to the m-th AP can be expressed as Based on the spherical wavefront propagation model, the near-field LoS channel (i.e., the channel matrix) from the m-th AP to the user can be modeled as:

[0046]

[0047] where is the complex path gain of the m-th AP, and b m (r, θ) represents the near-field steering vector of the m-th AP, that is:

[0048]

[0049] where represents the distance between the n-th antenna of the m-th AP and the user. After the above analysis, the channel matrix can be expressed in the following form:

[0050]

[0051] Since near-field communication needs to consider not only the angular domain but also the distance domain, the beam training method in far-field communication cannot be directly applied to near-field communication.

[0052] In the present invention, first, the relative direction of the user is estimated, that is, each AP estimates the relative direction of the user based on far-field beam training.

[0053] Specifically, the hierarchical beam training based on the angular domain is as follows Figure 2As shown, for the m-th AP, the codewords of each layer in the far-field codebook should cover a specific area. According to the idea of the dichotomy method, the coverage range of the upper-layer codewords is the union of the coverage ranges of the lower-layer codewords. For the codewords of the l-th layer, at the center of the active array, N l = 2 l antennas, the smaller N l is, the greater the probability that the user is located in the far field of this AP. There are a total of N l codewords in the l-th layer, and each codeword points to Let denote the set of codewords of the l-th layer of the m-th AP, then there is:

[0054]

[0055] where represents the i-th codeword of the l-th layer of the m-th AP, where b(∞,θ) = a(θ) represents the far-field steering vector,

[0056] Using the beam search based on the binary tree, the hierarchical beam training of the angular domain of the m-th AP can be realized. Assume represents the optimal codeword of the l-th layer of the m-th AP, and i * is the optimal codeword index. The angle of the user relative to the m-th AP can be estimated as while the actual user is located within the angular range of . Thus, each layer of training requires 2 training overheads (i.e., the occupied time resources). Therefore, after L layers of training for the m-th AP, the total overhead is 2L.

[0057] Through the above method, the beam direction of each AP for the user can be obtained, and at the same time, the positions of the known APs. Next, it is necessary to solve the intersection point generated by the two beam directions, and finally substitute the steering vector to deduce the optimal near-field codeword.

[0058] In one embodiment, since both the beam direction and the AP position are in the service area of the distributed MIMO system, belonging to the straight line (i.e., the beam direction) and the point (i.e., the AP position and the user position) in the service area coordinate system, therefore, the position of the user can be screened according to the geometric relationship of several beam directions.

[0059] Specifically, Figure 3(a) shows the beam directions of each obtained AP. In the figure, k1, k2, k3, and k4 respectively represent the slopes of the beam directions of 4 APs. Since the array beam direction pattern of the AP is symmetric within 360°, the identified directions may be symmetric about the array with a certain probability. Therefore, to improve accuracy, each AP uses two slopes that are opposite to each other. Thus, two straight lines can be obtained at each AP, that is, symmetric beam directions symmetric about the corresponding AP are generated according to the beam directions.

[0060] Next, it is necessary to select a beam direction or a symmetric beam direction for each AP as the final beam direction of the AP. As Figure 3 (b) shows, according to the method of permutation and combination, there are 2 M combinations for these straight lines. -k1, -k2, -k3, and -k4 respectively represent the slopes of the symmetric beam directions of 4 APs. That is to say, 2 M beam direction combinations are generated according to the beam directions or symmetric beam directions of each AP; where M represents the number of APs in the distributed MIMO system.

[0061] To screen out the most suitable set of straight line combinations, it is easy to know that the intersections generated by this set of straight lines that meet the conditions are denser. As Figure 3 (c) shows, the straight line combination that meets the conditions can be screened out by the minimum average distance between the intersections, that is, the optimal beam direction combination is selected according to the average distance between the intersections formed by the beam directions in the beam direction combination; the beam direction or symmetric beam direction is selected for each AP according to the optimal beam direction combination.

[0062] More specifically, selecting the optimal beam direction combination according to the average distance between the intersections formed by the beam directions in the beam direction combination includes: selecting the beam direction combination corresponding to the minimum value in the average distances as the optimal beam direction combination. Specifically, two different beam directions will generate intersections in the service area of the distributed MIMO system, and there will be a distance between any two intersections. The above-mentioned average distance refers to the average distance obtained after averaging all the distances. Then, each group of beam direction combinations will generate an average distance. The present invention selects the beam direction combination corresponding to the minimum value among multiple average distances as the final beam direction combination.

[0063] In the embodiment of the present invention, the minimum value in the average distances is selected based on the KD-Tree. The KD-Tree is an efficient multi-dimensional space data structure, especially suitable for nearest neighbor search and range query of low-dimensional data. By reasonably selecting the splitting axis and splitting point, a balanced tree structure can be constructed, thereby accelerating the query operation.

[0064] Then, a number of intersection points are determined according to a number of beam directions; based on the position information of each intersection point in the service area of the distributed MIMO system, the intersection points in the densely populated area are filtered out to form an intersection point set; the average position information is calculated based on the position information of the intersection points in the intersection point set; and the average position information is used as the position information of the user relative to the distributed MIMO system.

[0065] Since the user is far from some individual APs, the intersection point errors generated by the user and other lines are large. Therefore, in this embodiment, the density-based clustering algorithm is used to filter out the intersection points in the densely populated area. More specifically, the density-based DBSCAN clustering algorithm is selected to filter out some intersection points to form an intersection point set, as shown in Figure 3 (d).

[0066] Each intersection point in the intersection point set has a coordinate value in the service area coordinate system. Therefore, in this embodiment, the average value (r', θ') of these coordinate values is taken as the final user position information. Substituting it into the near-field steering vector can obtain the near-field optimal codeword of the user with respect to each AP, which can be expressed as:

[0067]

[0068] where v m =b(r', θ'), v m represents the optimal near-field codeword of the m-th AP.

[0069] In summary, the present invention mainly realizes the position estimation of the user through a series of calculations. The algorithm complexity is O(2 M-1 M 2 logM), where O(·) represents the complexity function. Assuming that the complexity of calculating the average distance of P intersection points by the KD-tree is PlogP, the number of intersection points of M lines is M(M - 1) / 2. Then there are 2 M groups. Substituting P = M(M - 1) / 2 and taking the approximation, the above algorithm complexity can be obtained. It can be found that when M is small, the algorithm complexity spent is acceptable.

[0070] The effectiveness of the method of the present invention is illustrated by the following simulation experiment, that is, the effectiveness of the method of the present invention is verified through MATLAB simulation.

[0071] Specifically, the cell radius is set to 100m, the number of APs is set to 4, which are evenly distributed on a circle with a radius of R. Each AP is equipped with 128 ULA antennas. Therefore, the total number of antennas is MN = 512. The wavelength λ is set to 0.01m. The Rayleigh distance of a single AP can reach 327m. The users are randomly distributed in the cell, and the achievable rate is R = log2(1 + γ|h H v| 2)It is calculated that h represents the channels of the user and all APs, and v = [v1,…,v m ,…,v M , where v m represents the steering vector of the m-th AP, γ is the signal-to-noise ratio, and all numerical results are obtained from more than 10,000 random positions.

[0072] As shown in Table 1, the training overheads of different beam training methods are compared. The far-field exhaustive beam training method needs to search MN angles; for the near-field exhaustive beam training method, the number of distance sampling points is S = 6, and its training overhead is the product of the angle sampling points and the distance sampling points. Therefore, the beam training overheads of the far-field exhaustive and near-field exhaustive methods are 512 and 3072 respectively.

[0073] For the near-field two-stage beam training method, which first searches in the angle domain and then in the distance domain, its training overhead is the sum of the angle sampling points and the distance sampling points, reaching 518; for the near-field two-stage hierarchical beam training, which uses a hierarchical method in the angle domain and the distance domain respectively, the training overhead is greatly reduced to 24.

[0074] In contrast, the method proposed in the present invention first obtains the angles of the user relative to different APs by using the far-field codebook, and multiple APs search simultaneously. The beam training overhead is only 2log2(N). Then, by using the two-screening method, the final estimated point of the user is obtained without generating additional training overhead. It can be found that the present invention fully exploits the advantages of this architecture of the distributed MIMO system, and the cooperation between multiple APs significantly reduces the beam training overhead.

[0075] Table 1

[0076]

[0077] In addition, Figure 4 shows the performance of the achievable rate versus the signal-to-noise ratio under different beam training methods, where the signal-to-noise ratio ranges from 5 dB to 15 dB and the AP deployment radius is set to 50 m. It can be found that the beam training method proposed in the present invention is superior to the two-stage near-field beam training method under centralized MIMO (i.e., the near-field two-stage beam training method in Table 1) and the two-stage hierarchical near-field beam training method (i.e., the near-field two-stage hierarchical beam training in Table 1), and is close to the near-field extreme domain codebook with a beam training overhead of thousands under centralized MIMO (i.e., the near-field exhaustive search beam training method in Table 1), and far superior to the performance of the centralized MIMO far-field codebook (i.e., the far-field exhaustive search beam training method in Table 1).

[0078] The simulation results show that the beam training method of the present invention can achieve an achievable rate close to that of the near-field extreme domain exhaustive beam training while significantly reducing the training overhead.

[0079] The present invention also discloses a near - field beam training device for a single - user distributed MIMO system, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the above - mentioned method is implemented.

[0080] Based on such understanding, all or part of the processes in the methods of the above - mentioned embodiments of the present invention can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a computer - readable storage medium. When the computer program is executed by the processor, the steps of the above - mentioned method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file, or some intermediate form, etc. The computer - readable medium can at least include: any entity or device capable of carrying the computer program code to a storage device, a recording medium, a computer memory, a read - only memory (ROM), a random - access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium. For example, a USB flash drive, a mobile hard disk, a magnetic disk, or an optical disc, etc. In some jurisdictions, according to legislation and patent practice, the computer - readable medium cannot be an electrical carrier signal and a telecommunication signal.

[0081] In the above - mentioned embodiments, the descriptions of each embodiment have their own focuses. For the parts not detailed or recorded in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0082] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed in the present invention can be implemented by electronic hardware, or by a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present invention.

[0083] In the embodiments provided by the present invention, it should be understood that the disclosed device / equipment and method can be implemented in other ways. For example, the device / equipment embodiments described above are merely illustrative.

[0084] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A near-field beam training method for a single-user distributed MIMO system, characterized in that, Including the following steps: Based on the far-field hierarchical codebooks of each AP in the distributed MIMO system, perform far-field hierarchical beam training on the corresponding AP respectively to obtain the beam directions of each AP pointing to the user; According to a plurality of the beam directions, screen the position of the user to obtain the position information of the user relative to the distributed MIMO system.

2. The near-field beam training method for a single-user distributed MIMO system according to claim 1, wherein After obtaining the position information of the user relative to the distributed MIMO system, it further includes: Calculate the near-field optimal codeword corresponding to each AP based on the position information.

3. The near-field beam training method of a single-user distributed MIMO system according to claim 1, characterized in that, According to a plurality of the beam directions, screening the position of the user includes: Screen the position of the user according to the geometric relationship of a plurality of the beam directions.

4. The near-field beam training method for a single-user distributed MIMO system according to claim 2 or 3, characterized in that, According to the geometric relationship of a plurality of the beam directions, screening the position of the user includes: Determine a plurality of intersection points according to a plurality of the beam directions; Based on the position information of each of the intersection points in the service area of the distributed MIMO system, screen out the intersection points in the position dense area to form an intersection point set; Calculate the average position information based on the position information of the intersection points in the intersection point set; Use the average position information as the position information of the user relative to the distributed MIMO system.

5. The near-field beam training method for a single-user distributed MIMO system according to claim 4, characterized in that, Before determining a plurality of intersection points according to a plurality of the beam directions, it further includes: Generate symmetric beam directions symmetric to the corresponding AP according to the beam directions; Select the beam direction or the symmetric beam direction for each AP as the final beam direction of the AP.

6. The near-field beam training method for a single-user distributed MIMO system according to claim 5, characterized in that Selecting the beam direction or the symmetric beam direction for each AP array includes: Generate 2 beam direction combinations according to the beam direction or symmetric beam direction of each AP; where M represents the number of APs in the distributed MIMO system; M two beam direction combinations; where M represents the number of APs in the distributed MIMO system; Select the optimal beam direction combination according to the average distance between the intersection points formed by the beam directions in the beam direction combination; Select the beam direction or the symmetric beam direction for each AP according to the optimal beam direction combination.

7. The near-field beam training method for a single-user distributed MIMO system according to claim 6, characterized in that Selecting the optimal beam direction combination according to the average distance between the intersection points formed by the beam directions in the beam direction combination includes: Select the beam direction combination corresponding to the minimum value in the average distance as the optimal beam direction combination.

8. The near-field beam training method of a single-user distributed MIMO system according to claim 7, characterized in that, Selecting the beam direction combination corresponding to the minimum value in the average distance as the optimal beam direction combination includes: Select the minimum value in the average distance based on the KD-Tree.

9. The near-field beam training method for a single-user distributed MIMO system according to claim 4, wherein Use the density-based clustering algorithm to screen out the intersection points in the position dense area.

10. A near-field beam training device for a single-user distributed MIMO system, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method according to any one of claims 1-9.

Images