A fast near-field beam training method for ultra-large-scale MIMO based on wavenumber domain spreading width

Abstract

The present invention provides a method for ultra-large-scale MIMO fast near-field beam training based on wavenumber domain diffusion width. The method comprises: obtaining an angle domain channel response between a base station array and a user terminal based on a far-field codebook; interpolating the angle domain channel response based on a discrete Fourier function to obtain a wavenumber domain channel response; performing diffusion interval detection in the wavenumber domain based on the wavenumber domain channel response to obtain a wavenumber domain diffusion interval; obtaining an angle estimate of the first angle based on a first functional relationship between a first angle between the user terminal and the base station array and the diffusion interval; obtaining a distance estimate of the first distance based on a second functional relationship between a first distance between the user terminal and the base station array and the diffusion interval; and obtaining a target codebook between the base station array and the user terminal based on the angle estimate and the distance estimate. Using the embodiments of the present invention for near-field beam training can reduce the time complexity of training, improve calculation accuracy, and accelerate processing speed.

Description

Technical Field

[0001] The present invention relates to the field of beam training technology, and in particular to a very large-scale MIMO fast near-field beam training method based on wavenumber domain diffusion width. Background Art

[0002] Extremely large-scale array (XL-array) is a key technology in the sixth generation mobile communication system (6G). By deploying an extremely large antenna array at the base station (BS), higher spectrum efficiency and energy efficiency are expected to be achieved. As the array aperture increases and the carrier frequency increases, the near-field area will expand, and users will be more likely to appear in the near-field area. Under near-field conditions, it is necessary to use the spherical wave assumption to calculate the user's channel response, rather than the plane wave assumption under far-field conditions. The user channel correlation calculated based on the spherical wave assumption is different from that under far-field conditions. A significant difference is that when two users are located at the same angle and different distances, the channel response calculated based on the far-field plane wave assumption is exactly the same, and the array will not be able to distinguish between the two users. However, the channel response correlation calculated based on the near-field spherical wave assumption will gradually decrease as the distance difference increases, and the channels of the two users will show certain differences.

[0003] Currently, the main fast beam training scheme is based on a far-field angle-domain codebook. This scheme directly transmits the far-field beamforming codebook to the base station and estimates the user angle and distance using the angle-domain channel response. However, this scheme has two drawbacks: First, the estimation algorithm lacks a closed-form solution and relies on an exhaustive search algorithm, resulting in high time complexity. Second, the angle-domain parameter estimation is performed on a series of discrete angle values, which can lead to a certain error between the estimated value and the actual value. Summary of the Invention

[0004] The purpose of the present invention is to provide a fast near-field beam training method for ultra-large-scale MIMO based on wavenumber domain diffusion width, which is used to solve the problems of high time complexity overhead, high error and low processing speed of the near-field beam training method in the prior art.

[0005] To solve the above technical problems, an embodiment of the present invention provides a method for fast near-field beam training of ultra-large-scale MIMO based on wavenumber domain spreading width, comprising:

[0006] Obtaining the angle domain channel response between the base station array and the user terminal based on the far-field beamforming codebook;

[0007] interpolating the angle domain channel response according to a discrete Fourier function to obtain a wavenumber domain channel response;

[0008] performing diffusion interval detection in the wavenumber domain according to the wavenumber domain channel response to obtain the diffusion interval in the wavenumber domain;

[0009] Obtaining an angle estimate of the first angle based on a first functional relationship between a first angle between the user terminal and the base station array and the diffusion interval and the diffusion interval; wherein the first functional relationship is obtained based on a stationary phase principle and the wavenumber domain channel response;

[0010] Obtaining a distance estimate of the first distance based on a second functional relationship between a first distance between the user terminal and the base station array and the diffusion interval and the diffusion interval, wherein the second functional relationship is obtained based on a stationary phase principle and the wavenumber domain channel response;

[0011] A target codebook between the base station array and the user terminal is obtained according to the angle estimation value and the distance estimation value.

[0012] Optionally, interpolating the angle domain channel response according to a discrete Fourier function to obtain a wavenumber domain channel response includes:

[0013] Obtaining a sampling position and a sampling interval in the wavenumber domain according to a third functional relationship between the angle domain channel response and the wavenumber domain channel response;

[0014] According to the sampling position and the sampling interval, the angle domain channel response is interpolated into the wave number domain channel response by using a discrete Fourier function.

[0015] Optionally, the method further includes:

[0016] determining, according to a stationary phase principle, a first interval in which the power of the wavenumber domain channel response is concentrated, wherein a boundary value of the first interval is represented by the first angle and the first distance;

[0017] The first functional relationship and the second functional relationship are obtained according to the corresponding relationship between the boundary value of the first interval and the diffusion interval.

[0018] Optionally, the method further includes:

[0019] Acquire the far-field beamforming codebook sent by the base station, wherein each far-field beamforming codebook occupies one time domain symbol;

[0020] The step of obtaining an angle domain channel response between the base station array and the user terminal according to the far-field beamforming codebook includes:

[0021] The far-field beamforming codebook is arranged according to the time-domain symbols to generate the angle-domain channel response between the base station array and the user terminal.

[0022] An embodiment of the present invention further provides a device for ultra-large-scale MIMO fast near-field beam training based on wavenumber domain spreading width, comprising:

[0023] A first processing module is configured to obtain an angle domain channel response between the base station array and the user terminal according to a far-field beamforming codebook;

[0024] A first calculation module is configured to interpolate the angle domain channel response according to a discrete Fourier function to obtain a wavenumber domain channel response;

[0025] A first detection module is configured to perform diffusion interval detection in the wavenumber domain according to the wavenumber domain channel response to obtain the diffusion interval in the wavenumber domain;

[0026] a second calculation module, configured to obtain an angle estimate of the first angle based on a first functional relationship between a first angle between the user terminal and the base station array and the diffusion interval, and the diffusion interval; wherein the first functional relationship is obtained based on a stationary phase principle and the wavenumber domain channel response;

[0027] a third calculation module, configured to obtain a distance estimate of the first distance based on a second functional relationship between the first distance between the user terminal and the base station array, the diffusion interval, and the diffusion interval, wherein the second functional relationship is obtained based on a stationary phase principle and the wavenumber domain channel response;

[0028] A fourth calculation module is configured to obtain a target codebook between the base station array and the user terminal according to the angle estimation value and the distance estimation value.

[0029] An embodiment of the present invention also provides a network device, comprising: a processor, a memory, and a program stored on the memory and runnable on the processor, wherein when the program is executed by the processor, the program implements the ultra-large-scale MIMO fast near-field beam training method based on wavenumber domain diffusion width as described in any one of the above items.

[0030] An embodiment of the present invention also provides a readable storage medium, comprising: a program stored on the readable storage medium, and when the program is executed by a processor, the steps of the ultra-large-scale MIMO fast near-field beam training method based on wavenumber domain diffusion width as described in any of the above items are implemented.

[0031] An embodiment of the present invention also provides a computer program product, including computer instructions, which, when executed by a processor, implement the steps of the ultra-large-scale MIMO fast near-field beam training method based on wavenumber domain diffusion width as described in any of the above items.

[0032] At least one of the above technical solutions of the present invention has the following beneficial effects:

[0033] In the above scheme, first, the far-field beamforming codebook is scanned to obtain the angle domain channel response between the base station array and the user terminal. Then, the angle domain channel response is interpolated according to the discrete Fourier function to obtain the wavenumber domain channel response. Secondly, based on the pre-analyzed first functional relationship between the spreading interval and the first angle and the second functional relationship between the spreading interval and the second angle, an angle estimate and a distance estimate are obtained. Finally, a target codebook is generated based on the angle estimate and the distance estimate. In the embodiment of the present invention, only the far-field beamforming codebook is used for search during near-field beam training, which significantly reduces the time complexity of training. In addition, the discrete angle domain channel response is restored to the continuous wavenumber domain channel response according to the interpolation operation of the discrete Fourier function. This can overcome the problem of errors in the angle estimate and distance estimate estimated in the angle domain, improve the accuracy of the angle estimate and distance estimate, and pre-analyze the relationship between the spreading interval, the first angle, and the first distance to obtain the angle estimate and distance estimate, which helps to reduce the time complexity of the algorithm and speed up the algorithm processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 Schematic diagram of the structure of a very large-scale array system according to an embodiment of the present invention;

[0035] Figure 2 Schematic diagram of a process for ultra-large-scale multiple-input multiple-output (MIMO) fast near-field beam training based on wavenumber domain spreading width according to an embodiment of the present invention;

[0036] Figure 3 Schematic diagram of the normalized mean square error result of angle estimation corresponding to the first embodiment of the present invention;

[0037] Figure 4 Schematic diagram of the normalized mean square error result of distance estimation corresponding to the first embodiment of the present invention;

[0038] Figure 5 This is a schematic diagram of the achievable rate results corresponding to the first embodiment of the present invention;

[0039] Figure 6 This is a schematic diagram of the effective achievable rate corresponding to the first embodiment of the present invention;

[0040] Figure 7 This is a structural diagram of a very large-scale multiple-input multiple-output (MIMO) fast near-field beam training device based on wavenumber domain diffusion width according to an embodiment of the present invention. DETAILED DESCRIPTION

[0041] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention and not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0042] An ultra-large-scale multiple-input multiple-output (MIMO) fast near-field beam training method based on wavenumber domain diffusion width according to an embodiment of the present invention is applied to a narrowband XL-array system, such as a system Figure 1 As shown, before describing the specific method of this application, the XL-array system is first introduced: the carrier frequency used by the system is f c , with the carrier wavelength denoted as λ. The BS uses a uniform linear array (ULA), with the antenna spacing in the ULA denoted as d = λ / 2. Therefore, the array aperture is D = (N-1)d, where the reference element is located at the center of the array. The origin of the coordinate axes is selected at the reference element, with the x-axis along the array. The distance from the user terminal to the reference antenna is r0, and the angle of departure (AOD) from the user terminal to the reference antenna is φ, denoted by Ω = cosφ.

[0043] Based on the above XL-array system, the beam training problem is modeled to obtain the initial model:

[0044] If the near-field range of the array is expressed in terms of the Rayleigh distance r ray Indicates: r ray =2D 2 / λ. Then when r0 <r ray When the user is located in the near-field area of ​​the BS, the downlink channel can be modeled using the near-field steering vector as:

[0045]

[0046] in, is the channel vector, h0 is the complex gain of the path at the reference antenna. According to the spherical wavefront propagation model, taking into account both amplitude and phase variations, the near-field steering vector can be modeled as:

[0047]

[0048] where r n represents the distance from the nth antenna of the BS array to the user terminal. The x coordinate of the nth antenna is in Then r n The calculation formula is

[0049] For the channel vector h near By performing a discrete Fourier transform (DFT), we can obtain the virtual angle domain representation of the channel, namely:

[0050]

[0051] where F=[a(Ω1),..,a(Ω N )]∈C N×N represents the Fourier transform matrix. represents the far-field steering vector, and

[0052] By adding x n Expanding to a continuous variable x, the near-field channel in formula (1) can be transformed into a A univariate function on , as follows:

[0053]

[0054] Performing Fourier transform (FT) on formula (4) can obtain the near-field channel response in the wavenumber domain, which is expressed as H near (k x ), where k x Indicates the wave number of the x-axis:

[0055]

[0056] like Figure 2 As shown, based on the above initial model, an embodiment of the present invention provides a super-large-scale multiple-input multiple-output (MIMO) fast near-field beam training method based on wavenumber domain diffusion width, including:

[0057] Step S201: obtaining an angle domain channel response between a base station array and a user terminal according to a far-field beamforming codebook;

[0058] In step S201, the far-field beamforming codebook is a(Ω n ), According to a(Ω n ) can be obtained as shown in formula (3): near,A .

[0059] Step S202, interpolating the angle domain channel response according to a discrete Fourier function to obtain a wavenumber domain channel response;

[0060] In step S202, since the wave number domain channel response H near (k x ) and the angle domain channel response H near,A They are the near-field channel response h near (x). Therefore, Hn ear,A Can be regarded as Hn ear (k x ) is sampled in the wavenumber domain, and hn ear Can be regarded as hn ear (x) is sampled in the spatial domain, and the sampling interval in the spatial domain is d = λ / 2. The wave number domain part supports communication power transmission outside the radiated near field, so the wave number domain bandwidth is recorded as because Satisfying the Nyquist sampling theorem, Hn ear,A Can be used to rebuild Hn ear (k x ). Therefore, according to the discrete Fourier function, the angle domain channel response Hn ear,A Interpolation is performed to obtain the wavenumber domain channel response Hn ear (k x ). Restoring the discrete angle domain channel response to the continuous wave number domain channel response can improve the calculation accuracy.

[0061] Step S203: performing diffusion interval detection in the wavenumber domain according to the wavenumber domain channel response to obtain the diffusion interval in the wavenumber domain;

[0062] In step S203, the diffusion interval is obtained according to the wavenumber domain channel response Wherein β is a given parameter. In actual use, considering the fluctuation characteristics of the wave number domain channel response, β is generally taken as 0.42, but the present invention is not limited to this.

[0063] Step S204: Obtain an angle estimate of the first angle based on a first functional relationship between a first angle between the user terminal and the base station array and the diffusion interval and the diffusion interval; wherein the first functional relationship is obtained based on a stationary phase principle and the wavenumber domain channel response;

[0064] In step S204, the first angle Ω is Ω=cosφ, where φ is the distance AOD from the user terminal to the reference antenna. The wave number domain channel response is processed according to the stationary phase principle to obtain an ideal expression of the diffusion interval. The ideal expression of the diffusion interval is represented by the first angle between the user terminal and the base station array and the first distance between the user terminal and the base station array. Therefore, it can be obtained that there is a first functional relationship between the diffusion interval and the first angle. After obtaining the diffusion interval in step S203, the angle estimation value of the first angle can be obtained based on the diffusion interval, which can effectively reduce the time complexity of the algorithm and speed up the algorithm processing speed.

[0065] Step S205: Obtaining a distance estimate of the first distance based on a second functional relationship between the first distance between the user terminal and the base station array, the diffusion interval, and the diffusion interval; wherein the second functional relationship is obtained based on a stationary phase principle and the wavenumber domain channel response;

[0066] In step S205, the first distance is the distance r0 from the user terminal to the reference antenna. By processing the wavenumber domain channel response according to the stationary phase principle, an ideal expression of the diffusion interval can be obtained. The ideal expression of the diffusion interval is represented by the first angle between the user terminal and the base station array and the first distance between the user terminal and the base station array. Therefore, it can be obtained that there is a second functional relationship between the diffusion interval and the first distance. Then, after obtaining the diffusion interval in step S203, a distance estimate of the first distance can be obtained based on the diffusion interval, which can effectively reduce the time complexity of the algorithm and speed up the algorithm processing.

[0067] Step S206: Obtain a target codebook between the base station array and the user terminal according to the angle estimation value and the distance estimation value.

[0068] In step S206, according to the angle estimation value and distance estimates Generate target codebook:

[0069] In an embodiment of the present invention, when performing near-field beam training, only the far-field beamforming codebook is used for searching, which significantly reduces the time complexity overhead of training. In addition, according to the interpolation operation of the discrete Fourier function, the discrete form of the angle domain channel response is restored to the continuous form of the wavenumber domain channel response, which can overcome the problem of errors in the angle estimation value and the distance estimation value estimated in the angle domain, improve the accuracy of the angle estimation value and the distance estimation value, and pre-analyze the relationship between the diffusion interval and the first angle and the first distance, thereby obtaining the angle estimation value and the distance estimation value, which helps to reduce the time complexity of the algorithm and speed up the algorithm processing speed.

[0070] Optionally, interpolating the angle domain channel response according to a discrete Fourier function to obtain a wavenumber domain channel response includes:

[0071] Obtaining a sampling position and a sampling interval in the wavenumber domain according to a third functional relationship between the angle domain channel response and the wavenumber domain channel response;

[0072] According to the sampling position and the sampling interval, the angle domain channel response is interpolated into the wave number domain channel response by using a discrete Fourier function.

[0073] In the embodiment of the present invention, a specific method for interpolating the angle domain channel response to obtain the wavenumber domain channel response in step S202 is described below:

[0074] By comparing the kernel function of formula (5) with the exponential component of formula (3), the sampling position in the wavenumber domain can be determined as The sampling interval in the wavenumber domain is expressed as Therefore, the wavenumber domain channel response Hn ear (k x ) can be obtained from the angle domain channel response Hn ear,A Interpolation is:

[0075]

[0076] in, express The nth element of It's Hn ear,A The period extension of and

[0077] Optionally, the method further includes:

[0078] determining, according to a stationary phase principle, a first interval in which the power of the wavenumber domain channel response is concentrated, wherein a boundary value of the first interval is represented by the first angle and the first distance;

[0079] The first functional relationship and the second functional relationship are obtained according to the corresponding relationship between the boundary value of the first interval and the diffusion interval.

[0080] In this embodiment of the present invention, a method for determining a first functional relationship between a first angle between a user terminal and a base station array and a diffusion range in step S204, and a second functional relationship between a first distance between the user terminal and the base station array and a diffusion range in step S205 is described.

[0081] According to the stationary phase principle, it can be determined that the power of the wavenumber domain channel response shown in the above formula (5) is mainly concentrated in the first interval Where, the diffusion interval is the actual detection value corresponding to the first interval. Specifically:

[0082] First, substitute the above formula (4) into formula (5), and we can get:

[0083]

[0084] The above formula (7) has the basic form of oscillatory integral, that is, I = ∫A(x)e jψ(x) According to the stationary phase principle, if the phase function ψ(x) has a stationary point x in the integral interval s ,Right now Then the oscillation integral I can be approximated as: If the phase function ψ(x) has no stationary point within the integration interval, the oscillation integral I can be approximated to 0. Therefore, if we want to study the specific location of the diffusion interval, we need to discuss when the phase function ψ(x) has a stationary point within the integration interval.

[0085] From formula (7), the expression of the phase function is: The first-order derivative is: Therefore the stationary point is:

[0086]

[0087] Since the integral range of formula (7) is [-D / 2, D / 2], when x s ∈[-D / 2,D / 2], we can get k x Need to belong to a first interval That is the theoretical expression of the diffusion range mentioned above. Substitute formula (8) into x s ∈[-D / 2,D / 2], we can get the first interval as:

[0088]

[0089] Among them, the first interval is the ideal expression of the diffusion interval.

[0090] When the user's first distance r0 is greater than the order of magnitude of the array aperture D, that is, r0>>D, the approximate simplified form of formula (9) can be obtained:

[0091]

[0092] According to the actual detected diffusion range And formula (10), let the boundary values ​​of the two intervals be equal, we can get the first functional relationship and the second functional relationship

[0093] Furthermore, the boundary value k of the diffusion interval is l , and k r Substitute the first functional relationship and the angle can be estimated The boundary value k of the diffusion interval l , and k r , and the angle estimate Substituting the second function relationship, we can get the distance estimate

[0094] The closed-form relationship between the diffusion interval, the first angle, and the first distance obtained according to the stationary phase principle can reduce the time complexity of the estimation algorithm.

[0095] Optionally, the method further includes:

[0096] Acquire the far-field beamforming codebook sent by the base station, wherein each far-field beamforming codebook occupies one time domain symbol;

[0097] The step of obtaining an angle domain channel response between the base station array and the user terminal according to the far-field beamforming codebook includes:

[0098] The far-field beamforming codebook is arranged according to the time-domain symbols to generate the angle-domain channel response between the base station array and the user terminal.

[0099] In the embodiment of the present invention, the far-field beamforming codebook a(Ω) sent by the BS is obtained. n ), n∈ Each codebook occupies one time domain symbol; then, according to the time domain symbol, the far-field beamforming codebook is arranged to generate the angle domain channel response between the base station array and the user terminal. Specifically, at the nth time domain symbol, the received signal is a H (Ω n )·h near , the signals received by each symbol are arranged into a vector, and the angle domain channel response H can be obtained near,A . It can effectively reduce the time overhead of beam training.

[0100] The following combination Figures 3 to 6 It should be noted that the following examples are only used to illustrate the present invention and are not intended to limit the present invention.

[0101] Example 1: A simulation test is performed using the ultra-large-scale MIMO fast near-field beam training method based on wavenumber domain spreading width provided by the present invention. The specific process is as follows:

[0102] In the simulation, the carrier frequency is set to f c =30GHz, the ULA has N=256 antennas, and the antenna spacing is λ / 2, so the array aperture is Therefore, the Rayleigh distance of ULA is r ray 1000 sets of random user channels are also set, where the first distance between the user terminal and the reference antenna is set to r0 = 20 m, and the first angle Ω between the user terminal and the reference antenna is set to a uniformly distributed random number distributed in [-1, 1].

[0103] The simulation is performed under different signal-to-noise ratios, i.e., the received signal y of the i-th random user under the n-th signal-to-noise ratio is i is defined as: in, The signal-to-noise ratio is defined as:

[0104] In order to compare the differences between the ultra-large-scale MIMO fast near-field beam training method based on wavenumber domain spread width provided by the present invention and existing solutions, two indicators, the normalized mean square error (NMSE) of the estimation error and the achievable rate, are used as comparison criteria. Among them, the existing solutions used for comparison in the embodiments of the present invention include a beam training method based on an exhaustive search algorithm, a beam training method based on an angle support width joint angle and range estimation algorithm (ASW-JE) (the parameter K is set to 3), and a method for calculating the performance upper bound of the optimal beam through target-accurate channel state information (CSI).

[0105] First, we compare the error of the estimation algorithm. For the angle estimation value of the first angle, its NMSE is defined as: For the distance estimate of the first distance, its NMSE is defined as:

[0106] The simulation results are as follows: Figure 3 and Figure 4 As shown, Figure 3The NMSE results of the angle estimation values ​​of the ultra-large-scale MIMO fast near-field beam training method based on wavenumber domain spread width provided by the present invention and the beam training method based on exhaustive search algorithm and ASW-JE (K=3) in the existing scheme are shown. Figure 4 The NMSE results of the distance estimation values ​​of the ultra-large-scale MIMO fast near-field beam training method based on wavenumber domain diffusion width provided by the present invention and the beam training method based on exhaustive search algorithm and ASW-JE (K=3) in the existing scheme are shown.

[0107] The results show that under high signal-to-noise ratio, the ultra-large-scale MIMO fast near-field beam training method based on wavenumber domain diffusion width provided by the present invention can obtain lower estimation error than the existing scheme, and when estimating the angle, due to the use of wavenumber domain interpolation technology, the error of the angle estimation can be lower than that of the exhaustive search algorithm.

[0108] The achievable rate metrics are then compared via simulation:

[0109] The achievable rate is defined as: Where v is the optimal beam obtained by each beam training method, Represents the real channel vector. In order to more specifically compare the advantages and disadvantages of different methods in terms of beam training overhead, we further consider the effective achievable rate indicator, which is defined as:

[0110]

[0111] Where T tra represents the time domain symbol overhead of each method, T of the exhaustive search algorithm tra =256*5=1280, T of ASW-JE (K=3) tra =256+3=259, the T of the method proposed in the present invention tra =256. Take the total number of symbols in the time domain as T tot =10000.

[0112] According to the achievable rate and effective achievable rate, simulation is performed, and the simulation results are as follows Figure 5 and Figure 6 As shown, Figure 6 The simulation results of the achievable rates using the ultra-large-scale MIMO fast near-field beam training method based on wavenumber domain spreading width provided by the present invention, the beam training method based on exhaustive search algorithm in existing solutions, ASW-JE (K=3), and the performance upper bound method for calculating the optimal beam with precise CSI are shown. Figure 4The simulation results of the effective achievable rate using the ultra-large-scale MIMO fast near-field beam training method based on wavenumber domain diffusion width provided by the present invention and the beam training method based on exhaustive search algorithm in the existing scheme, ASW-JE (K=3), and the performance upper bound method of the optimal beam calculation with precise target CSI are shown.

[0113] The simulation results show that, considering the achievable rate, the proposed method outperforms the exhaustive search algorithm at high signal-to-noise ratios because it has lower angle estimation error. Furthermore, considering the algorithm overhead, the proposed method significantly outperforms the exhaustive search algorithm and the ASW-JE (K=3) method in terms of effective achievable rate due to its lower training overhead.

[0114] like Figure 7 As shown, an embodiment of the present invention further provides a very large-scale MIMO fast near-field beam training device based on wavenumber domain spreading width, comprising:

[0115] A first processing module 701 is configured to obtain an angle domain channel response between a base station array and a user terminal according to a far-field beamforming codebook;

[0116] A first calculation module 702 is configured to interpolate the angle domain channel response according to a discrete Fourier function to obtain a wavenumber domain channel response;

[0117] A first detection module 703 is configured to perform a diffusion interval detection in the wavenumber domain according to the wavenumber domain channel response to obtain the diffusion interval in the wavenumber domain;

[0118] A second calculation module 704 is configured to obtain an angle estimate of the first angle based on a first functional relationship between a first angle between the user terminal and the base station array and the diffusion interval and the diffusion interval, wherein the first functional relationship is obtained based on a stationary phase principle and the wavenumber domain channel response;

[0119] A third calculation module 705 is configured to obtain a distance estimate of the first distance based on a second functional relationship between the first distance between the user terminal and the base station array, the diffusion interval, and the diffusion interval, wherein the second functional relationship is obtained based on a stationary phase principle and the wavenumber domain channel response;

[0120] The fourth calculation module 706 is configured to obtain a target codebook between the base station array and the user terminal according to the angle estimation value and the distance estimation value.

[0121] Optionally, the first calculation module 702 includes:

[0122] a first calculation unit, configured to obtain a sampling position and a sampling interval in the wavenumber domain according to a third functional relationship between the angle domain channel response and the wavenumber domain channel response;

[0123] The second calculation unit is configured to interpolate the angle domain channel response into the wavenumber domain channel response through a discrete Fourier function according to the sampling position and the sampling interval.

[0124] Optionally, the device further comprises:

[0125] A first determining module is configured to determine, based on a stationary phase principle, a first interval in which the power of the wavenumber domain channel response is concentrated, wherein a boundary value of the first interval is represented by the first angle and the first distance;

[0126] A fifth calculation module is configured to obtain the first functional relationship and the second functional relationship according to a correspondence between a boundary value of the first interval and the diffusion interval.

[0127] Optionally, the device further comprises:

[0128] A first acquisition module is configured to acquire the far-field beamforming codebook sent by a base station, wherein each far-field beamforming codebook occupies one time domain symbol;

[0129] The first processing module 701 includes:

[0130] A first arrangement unit is configured to arrange the far-field beamforming codebook according to the time domain symbols to generate the angle domain channel response between the base station array and the user terminal.

[0131] It should be noted that the embodiment of the device is a device corresponding to the embodiment of the above method, and all implementation methods in the embodiment of the above method are applicable to the embodiment of the device and can achieve the same technical effect.

[0132] An embodiment of the present invention also provides a network device, comprising: a processor, a memory, and a program stored on the memory and runnable on the processor. When the program is executed by the processor, the program implements the ultra-large-scale MIMO fast near-field beam training method based on wavenumber domain diffusion width as described in any of the above items, and can achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0133] An embodiment of the present invention further provides a computer-readable storage medium, comprising: a program stored on the computer-readable storage medium, wherein when the program is executed by a processor, the program implements the steps of the ultra-large-scale MIMO fast near-field beam training method based on wavenumber domain spread width as described in any of the above items, and can achieve the same technical effect. To avoid repetition, the details are not described here. The computer-readable storage medium is, for example, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0134] An embodiment of the present invention also provides a computer program product, including computer instructions. When the computer instructions are executed by a processor, the steps of the ultra-large-scale MIMO fast near-field beam training method based on wavenumber domain diffusion width as described in any of the above items are implemented, and the same technical effect can be achieved. To avoid repetition, they are not repeated here.

[0135] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "include", "comprises" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or terminal device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprising a ..." does not exclude the presence of other identical elements in the process, method, article or device comprising the element.

[0136] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if such changes and modifications fall within the scope of the claims and their equivalents, the present invention is intended to include such changes and modifications.

Claims

1. A fast near-field beam training method for ultra-large-scale MIMO based on wavenumber domain spreading width, characterized in that: include: Obtaining the angle domain channel response between the base station array and the user terminal based on the far-field beamforming codebook; interpolating the angle domain channel response according to a discrete Fourier function to obtain a wavenumber domain channel response; performing diffusion interval detection in the wavenumber domain according to the wavenumber domain channel response to obtain the diffusion interval in the wavenumber domain; Obtaining an angle estimate of the first angle based on a first functional relationship between a first angle between the user terminal and the base station array and the diffusion interval and the diffusion interval; wherein the first functional relationship is obtained based on a stationary phase principle and the wavenumber domain channel response; Obtaining a distance estimate of the first distance based on a second functional relationship between a first distance between the user terminal and the base station array and the diffusion interval and the diffusion interval, wherein the second functional relationship is obtained based on a stationary phase principle and the wavenumber domain channel response; A target codebook between the base station array and the user terminal is obtained according to the angle estimation value and the distance estimation value.

2. The ultra-large-scale MIMO fast near-field beam training method based on wavenumber domain spreading width according to claim 1, characterized in that: Interpolating the angle domain channel response according to a discrete Fourier function to obtain a wavenumber domain channel response includes: Obtaining a sampling position and a sampling interval in the wavenumber domain according to a third functional relationship between the angle domain channel response and the wavenumber domain channel response; According to the sampling position and the sampling interval, the angle domain channel response is interpolated into the wave number domain channel response by using a discrete Fourier function.

3. The ultra-large-scale MIMO fast near-field beam training method based on wavenumber domain spreading width according to claim 1, characterized in that: The method further comprises: determining, according to a stationary phase principle, a first interval in which the power of the wavenumber domain channel response is concentrated, wherein a boundary value of the first interval is represented by the first angle and the first distance; The first functional relationship and the second functional relationship are obtained according to the corresponding relationship between the boundary value of the first interval and the diffusion interval.

4. The ultra-large-scale MIMO fast near-field beam training method based on wavenumber domain spreading width according to claim 1, characterized in that The method further comprises: Acquire the far-field beamforming codebook sent by the base station, wherein each far-field beamforming codebook occupies one time domain symbol; The step of obtaining an angle domain channel response between the base station array and the user terminal according to the far-field beamforming codebook includes: The far-field beamforming codebook is arranged according to the time-domain symbols to generate the angle-domain channel response between the base station array and the user terminal.

5. A fast near-field beam training device for ultra-large-scale MIMO based on wavenumber domain spreading width, characterized in that: include: A first processing module is configured to obtain an angle domain channel response between the base station array and the user terminal according to a far-field beamforming codebook; A first calculation module is configured to interpolate the angle domain channel response according to a discrete Fourier function to obtain a wavenumber domain channel response; A first detection module is configured to perform diffusion interval detection in the wavenumber domain according to the wavenumber domain channel response to obtain the diffusion interval in the wavenumber domain; a second calculation module, configured to obtain an angle estimate of the first angle based on a first functional relationship between a first angle between the user terminal and the base station array and the diffusion interval, and the diffusion interval; wherein the first functional relationship is obtained based on a stationary phase principle and the wavenumber domain channel response; a third calculation module, configured to obtain a distance estimate of the first distance based on a second functional relationship between the first distance between the user terminal and the base station array, the diffusion interval, and the diffusion interval, wherein the second functional relationship is obtained based on a stationary phase principle and the wavenumber domain channel response; A fourth calculation module is configured to obtain a target codebook between the base station array and the user terminal according to the angle estimation value and the distance estimation value.

6. A network device, characterized in that: include: A processor, a memory, and a program stored on the memory and executable on the processor, wherein the program, when executed by the processor, implements the ultra-large-scale MIMO fast near-field beam training method based on wavenumber domain diffusion width as described in any one of claims 1 to 4.

7. A readable storage medium, characterized in that: include: The readable storage medium stores a program, and when the program is executed by the processor, the steps of the ultra-large-scale MIMO fast near-field beam training method based on wavenumber domain spread width are implemented as described in any one of claims 1 to 4.

8. A computer program product, characterized in that The method comprises computer instructions, which, when executed by a processor, implement the steps of the ultra-large-scale MIMO fast near-field beam training method based on wavenumber domain diffusion width as described in any one of claims 1 to 4.

Images