An effective method for near-field parameter estimation and positioning in XL-MIMO systems

Abstract

The present invention relates to an effective method for near-field parameter estimation and positioning in an XL-MIMO system. Considering a more challenging communication scenario where the LoS path is blocked and a system with extremely large-scale antenna arrays deployed at both the transmitter and receiver, the present invention first utilizes the structural relationship between the codebook matrix and near-field parameters to achieve preliminary extraction of the parameters. Subsequently, the steepest descent idea is used to iteratively refine the obtained parameters, thereby achieving super-resolution estimation of parameters such as angles, distances, and path gains. The obtained parameters can also be further applied to the sensing link. With the geometric constraints of the system and the obtained high-precision parameters, the present invention completes the accurate estimation of the location information of the user equipment and scattering points, providing a basis for the development of diverse sensing services. In addition, the proposed method is superior to existing competing methods in terms of both parameter estimation accuracy and positioning accuracy, and can still maintain excellent performance under the conditions of low sampling numbers and low pilot overhead.

Description

Technical Field

[0001] This technology relates to the field of wireless communication, and particularly to the estimation of near-field parameters in an XL-MIMO system and the positioning of key nodes in the system. Background Art

[0002] As a promising technology in the sixth-generation (6G) mobile communication, extremely large-scale MIMO (XL-MIMO) technology is expected to meet the growing traffic demand and overcome the severe path loss in high-frequency communication. Compared with traditional MIMO systems, the number of antennas in XL-MIMO systems increases significantly. At the same time, the use of high-frequency bands such as millimeter waves and terahertz makes the electromagnetic characteristic differences of wireless signals in different regions non-negligible. The communication area can be roughly divided into two regions, the near field and the far field, with the Rayleigh distance as the boundary. For XL-MIMO systems, with the increase in the number of deployed antennas and the working frequency, the Rayleigh distance can be up to hundreds of meters, which leads to a significant expansion of the near-field range.

[0003] Channel estimation algorithms based on the far-field plane wave assumption will experience performance degradation in the near-field region. The spherical wave assumption is widely used in near-field communication because it describes the information in both the angular dimension and the distance dimension and is closer to the real channel model. Under the spherical wave assumption, the near-field channel is determined by parameters such as angles, distances, and path gains. Therefore, implementing the estimation of channel parameters can not only reconstruct the near-field channel but also help to achieve the positioning of key nodes in the system.

[0004] In the paper by Z. Lu, Y. Han, S. Jin, and M. Matthaiou (Near-Field Localization and Channel Reconstruction for ELAA Systems [J]. IEEE Trans. Wireless Commun., vol. 23, no. 7, pp. 6938-6953, July 2024.), a damped Newtonized orthogonal matching pursuit method is proposed to estimate the parameters in the near-field channel. However, this method is only applicable to systems with single-antenna users, and its application scope is limited. In the paper by Y. Lu and L. Dai (Near-Field Channel Estimation in Mixed LoS / NLoS Environments for Extremely Large-Scale MIMO Systems [J]. IEEE Trans. Commun, vol. 71, no. 6, pp. 3694-3707, June 2023.), the channel parameters can be solved by using the polarization-domain orthogonal matching pursuit (PSOMP) algorithm. However, the parameter estimation accuracy of this method needs to be further improved. Summary of the Invention

[0005] Object of the Invention: To overcome the deficiencies of the prior art, the present invention proposes an effective near-field parameter estimation and positioning method for XL-MIMO systems to achieve accurate estimation of near-field parameters and accurate positioning of user equipment and environmental scatterers.

[0006] Technical Solution: An effective near-field parameter estimation and positioning method for XL-MIMO systems according to the present invention includes:

[0007] In a near-field communication scenario where the LoS path is blocked, for a system with ultra-large-scale uniform linear arrays deployed at both the transmitter and receiver ends, a spherical wave hypothesis is used to construct a near-field channel and received signal model;

[0008] Design a sampling codebook matrix corresponding to the angle and distance parameters, and based on the correlation between the received signal and the codebook matrix, use the idea of the orthogonal matching pursuit algorithm to solve for the index information related to the near-field parameters;

[0009] Analyze the association between the obtained index and the parameter to be estimated, and based on the structural characteristics of the codebook matrix and the Kronecker inner product operation rule, realize the preliminary extraction of the angle parameter and the distance parameter;

[0010] Design an optimization function to perform super-resolution refinement on the obtained parameters, and use the idea of the steepest descent to iteratively update the angle and distance parameters;

[0011] Utilize the geometric constraints of the system and the estimated angle and distance parameters to sense the key nodes of the system, thereby achieving accurate positioning of the user equipment and environmental scatter points;

[0012] Furthermore, in a near-field communication scenario where the LoS path is blocked, for a system with ultra-large-scale uniform linear arrays deployed at both the transmitter and receiver ends, a spherical wave hypothesis is adopted to construct a near-field channel and received signal model, specifically including:

[0013] Consider a communication system operating in the millimeter-wave band. Due to the introduction of ultra-large-scale arrays and the use of high-frequency operating bands, the Rayleigh distance can reach the order of hundreds of meters, and the coverage range of the near-field region is significantly expanded. Both the user equipment and the scatter points are located within the near-field radiation region.

[0014] In the proposed system, both the base station side and the user equipment side are equipped with ultra-large-scale uniform linear arrays, and the number of antennas is and respectively. Although high-frequency signals are beneficial to improving the communication rate and spectral efficiency of the system, the severe path loss and blockage sensitivity characteristics brought by high-frequency electromagnetic waves make the LoS path between the base station and the user equipment easily blocked by a dense environment. This situation is common in reality and is also very challenging.

[0015] To cope with such challenging communication scenarios and achieve effective near-field communication and sensing, the present invention focuses on a near-field communication system where the LoS path is blocked and the communication between the transceiver ends is only achieved through the NLoS path.

[0016] Considering the electromagnetic wave characteristics of the near field, the present invention adopts a more accurate spherical wave model for channel modeling to better simulate the real communication channel. Compared with the traditional plane wave model, the spherical wave model adds a description of the distance dimension in addition to the angle dimension, providing a basis for simultaneously realizing communication and sensing. The corresponding expression of the channel is as follows:

[0017]

[0018] where, is the number of paths, is the path gain corresponding to the th path, . and respectively correspond to the departure angle and arrival angle in the th path. and is the distance between the center point of the transceiver array and the th scattering point. denotes the conjugate transpose operation, and are the steering vectors of the transceiver respectively. Under the spherical wave model, the steering vector of the antenna is not only related to the angle parameter but also affected by the distance parameter, and its expression is as follows:

[0019]

[0020]

[0021] where, is the carrier frequency, is the speed of light, and and are the distances between the th and th antennas of the base station and user equipment antenna arrays respectively and the th scattering point, and its expression is written as:

[0022]

[0023]

[0024] where, , .

[0025] For the convenience of subsequent processing, the matrix mode of the channel is given accordingly: . Where, , , .

[0026] Assume that the base station sends pilot signals, then the signal received at the user equipment can be written as:

[0027]

[0028] where, is the precoding matrix that satisfies the constant modulus constraint, is the combining matrix that satisfies the constant modulus constraint, is the corresponding received signal dimension, is the noise matrix that follows the Gaussian distribution.

[0029] Furthermore, a sampling codebook matrix corresponding to the design angle and distance parameters is designed. According to the correlation between the received signal and the codebook matrix, the index information related to the near-field parameters is obtained by using the idea of the orthogonal matching pursuit algorithm, specifically including:

[0030] Through uniform sampling, the sampling values of the given angle and distance are obtained, thus getting as the sampling value of the angle of departure, as the sampling value of the angle of arrival, as the sampling value of the distance between the center of the base station array and the scattering point, and as the sampling value of the distance between the center of the user equipment array and the scattering point, where respectively correspond to the angle and distance sampling numbers of the transceiver. For the convenience of subsequent parameter extraction, here we define , , , and .

[0031] Substitute the above parameter sampling information into the steering vector expression corresponding to the spherical wave model and combine them to design the codebook matrices and , and their expressions are as follows:

[0032]

[0033]

[0034] Among them, , .

[0035] By virtue of the correlation of the matrix structure, the corresponding index information can be obtained through the idea of greedy search. Specifically, first, an initial residual and an initial sensing matrix are given. The vector is a vector for storing the index obtained in each iteration, and the number of iterations is equal to the number of paths .

[0036] Define and , and using the designed codebook, the residual is sparsified by virtue of the idea of the orthogonal matching pursuit algorithm, so as to obtain the sparse representation of the residual matrix in the angular domain and the distance domain:

[0037]

[0038] Subsequently, vectorize , and solve for the index corresponding to the maximum value in the obtained vector, that is,

[0039]

[0040] Among them, , represents the th column of the matrix, and the symbol represents the Kronecker product, represents the conjugate operation, represents the vectorization operation.

[0041] After completing the above steps, can be updated accordingly. It can be analyzed that the obtained index carries information related to the target angle and distance parameters. Therefore, by analyzing the internal structural relationship between the codebook matrix and the channel matrix, the update of the sensing matrix can be realized, and the specific operations are as follows:

[0042]

[0043] Among them, , . represents the transpose operation, represents the ceiling operation.

[0044] Using the solved to reconstruct the sparse channel vector, and thus obtain:

[0045]

[0046] Among them, represents the inverse operation, mapping the obtained vector into matrix form, that is . Immediately afterwards, update the residual as follows:

[0047]

[0048] And repeat the above process until the maximum number of iterations is reached. At this point, the index vector and the rough estimate of the channel are obtained.

[0049] Furthermore, analyze the relationship between the obtained index and the parameter to be estimated, and according to the structural characteristics of the codebook matrix and the Kronecker product operation rules, realize the preliminary extraction of the angle parameter and the distance parameter, specifically including:

[0050] Denote , and it can be analyzed that each index stored in the vector points to the row in with the strongest correlation with the current residual. And contains all combinations of angle and distance sampling information and is calculated by the Kronecker product. This means Each row in it corresponds to a sampling value of a set of determined angle and distance parameters.

[0051] Therefore, by using the obtained index information and the operation rules of the Kronecker product, the preliminary extraction of the angle and distance parameters can be realized, so as to obtain , , , , where

[0052]

[0053]

[0054]

[0055]

[0056] After the preliminary extraction of the angle and distance parameters is completed, the corresponding path gain can be solved by the least squares idea, and the solving process is as follows:

[0057]

[0058] where , . So far, the present invention has obtained a preliminary estimate including the departure angle, the arrival angle, the distance between the center of gravity of the base station array and the scattering point, the distance between the center of gravity of the user equipment array and the scattering point, and the channel gain, that is . These parameter information not only helps the reconstruction of the near-field channel, but also can be applied to the sensing link to assist the system to realize the position estimation of the user equipment and the positioning of the scattering point.

[0059] Furthermore, design an optimization function to refine the obtained parameters with super-resolution, and use the steepest descent idea to iteratively update the angle and distance parameters, specifically including:

[0060] Using the maximum likelihood idea, design the optimization function as follows:

[0061]

[0062] where represents the operation of taking the F norm, , is the given regularization parameter, is a diagonal matrix, is the modulus operation, is a constant to ensure the logarithm function is meaningful.

[0063] Next, perform iterative optimization on the objective function, given the number of iterations , and taking the th iteration as an example for illustration. Since the near-field model additionally considers the distance dimension, the number of parameters to be updated in the optimization process has increased significantly. To reduce the computational burden of the optimization process, the present invention explores the connection between the optimization of path parameters and the optimization of angular distance parameters.

[0064] Specifically, taking the partial derivative of the optimization function with respect to can obtain the optimal solution of the path parameters, and is a function related to angles and distances, and its expression is as follows:

[0065]

[0066] where , , , .

[0067] Substituting the expression of into the optimization function can obtain:

[0068]

[0069] Next, taking obtained from the above preliminary extraction process as the initial value of the parameter refinement link, the following parameter update is performed:

[0070]

[0071]

[0072]

[0073]

[0074] where is the update step size, , , and are the partial derivatives of with respect to , , and respectively, corresponding to the gradient descent directions of the transceiver angle and distance parameters.

[0075] Specifically, taking the th path as an example, let , can be calculated by the following formula:

[0076]

[0077] Among them, , .

[0078] The expression of can be written as , where , .

[0079] Similarly, the distance update direction between the center of the base station array and the scattering point of the th path can be calculated by the following formula:

[0080]

[0081] where

[0082]

[0083]

[0084]

[0085] Since the expression of is similar to the expression of is similar, for the sake of simplicity of representation, it is omitted here.

[0086] After completing the update of by means of the steepest descent idea, the corresponding can be obtained. Loop this process until the iteration ends. Finally, the super-resolution estimation of the parameters is realized, and the final result is denoted as . Substituting the obtained parameters into the channel model, the final estimation of the near-field channel can be obtained. Thus, the present invention realizes the high-precision estimation of near-field parameters such as angle, distance and path gain and the accurate reconstruction of the near-field channel matrix.

[0087] Furthermore, by using the geometric constraints of the system and the estimated angle and distance parameters to sense the key nodes of the system, the accurate positioning of the user equipment and the environmental scattering points can be realized, specifically including:

[0088] Performing geometric analysis on the system and combining the obtained high-precision parameters with the geometric relationship between nodes, the positioning of the scattering points in the environment can be realized first. Specifically, the position of the scattering point corresponding to the th path can be calculated by the following formula, that is,

[0089]

[0090] After obtaining the scattering point positions corresponding to all paths, the obtained scattering point position information can be regarded as known information, thereby assisting in the positioning of the user equipment. The position solution formula of the user equipment is given as follows:

[0091]

[0092] So far, the present invention has achieved accurate estimation of the positions of the user equipment and the scattering points in the near-field environment.

[0093] Advantageous effects: Compared with the prior art, the main advantages of the present invention are as follows: For a system with extremely large-scale antenna arrays deployed at both the transceiver ends, an efficient near-field parameter estimation and sensing method is proposed. By making full use of the structural characteristics of the codebook matrix and the antenna steering vector, the present invention provides an effective idea for extracting near-field parameters, and refines the initially extracted parameters using the idea of the steepest descent, thereby achieving high-precision estimation of parameters such as angles, distances, and path gains. In addition, the obtained high-precision parameters can be further applied to the sensing link. With the geometric constraints of the system and the obtained parameter information, the present invention has achieved accurate estimation of the position information of the user equipment and the scattering points, providing a basis for the development of diverse sensing services. Brief Description of the Drawings

[0094] Figure 1 It is a flowchart of the method for implementing near-field parameter estimation and system key node positioning in the XL-MIMO system of the present invention;

[0095] Figure 2 It is a schematic structural diagram of the XL-MIMO system of the present invention;

[0096] Figure 3 It is a performance graph of the root mean square error (RMSE) of parameter estimation of the present invention and the existing PSOMP method under different signal-to-noise ratios (SNRs);

[0097] Figure 4 It is a performance graph of the positioning RMSE of the present invention and the existing PSOMP method under different SNRs;

[0098] Figure 5 It is for the present invention when SNR = 30 dB, different numbers of distance samples under the parameter estimation and positioning RMSE performance graph;

[0099] Figure 6 It is for the present invention when SNR = 10 dB, different numbers of pilot frequencies under the parameter estimation and positioning RMSE performance graph; Detailed Description of the Invention

[0100] To make the features and advantages of the present invention more obvious and understandable, and to more clearly and definitively define the protection scope of the present invention, the following will elaborate on the preferred embodiments of the present invention in conjunction with the accompanying drawings.

[0101] Figure 2 It is a schematic diagram of the XL-MIMO system structure of the present invention. As Figure 2 shown, consider an XL-MIMO system where both the transmitter and receiver use ultra-large-scale uniform linear arrays. Among them, both the base station and the user equipment are located in the near-field region, and are respectively equipped with and antenna elements. In this system, the LoS path is blocked, and the communication process is only completed relying on the NLoS path. The position of the base station is known, and the positions of the user equipment and the scatterers need to be estimated.

[0102] Embodiment 1

[0103] Refer to Figure 3 and Figure 4 , Figure 3 and Figure 4 are respectively the RMSE performance graphs of parameter estimation and positioning of the present invention and the existing PSOMP method under different SNRs. The system parameters are set as: , , , , , , , . Establish a coordinate system with the position of the base station as the origin, and given the position of the user equipment to be estimated, as well as the coordinates of the scatterer, , , . It can be found from Figure 3 and Figure 4 that the proposed method is always superior to the existing PSOMP algorithm in terms of the RMSE performance of near-field parameter estimation and the positioning of the user equipment and the scatterer. As the system SNR increases, the overall performance of the proposed method has a significant improvement, while the parameter estimation accuracy and positioning effect of the existing PSOMP algorithm are not ideal under medium and low signal-to-noise ratio conditions. When SNR = 15 dB, the parameter estimation accuracy of the proposed method can reach to order of magnitude, and centimeter-level positioning can be achieved in terms of positioning performance.

[0104] Embodiment 2

[0105] Refer to Figure 5 , Figure 5For the present invention, when SNR = 30 dB, the parameter estimation and positioning RMSE performance graph for different numbers of distance samples ; except for the number of distance samples All other system parameter settings are the same as in Example 1. In Example 2, we set the variation trend of the number of distance samples as: . From Figure 5 it can be observed that when the number of samples of the codebook matrix increases to a certain value, that is, Figure 5 in when, the parameter estimation accuracy of the proposed method reaches the lower bound and no longer changes with the increase of . This characteristic provides reference information for how to select an appropriate number of samples. In addition, Figure 5 also shows that even when the number of samples is small, the proposed method has good performance in near-field parameter estimation and positioning.

[0106] Implementation of Example 3

[0107] See Figure 6 , Figure 6 For the present invention, when SNR = 10 dB, the parameter estimation and positioning RMSE performance graph for different numbers of pilot frequencies ; except for the number of pilot frequencies All other system parameter settings are the same as in Example 1. The variation trend of the number of pilot frequencies in Example 3 is set as: . From Figure 6 it can be observed that as increases, both the parameter estimation and positioning performance of the proposed method are improved accordingly. In addition, it can also be seen from Figure 6 that the proposed method has an advantage in saving pilot overhead. Even under the condition of , the proposed method can still provide accurate parameter estimation and centimeter-level positioning services.

[0108] In summary, the present invention considers an effective method for near-field parameter estimation and positioning in an XL-MIMO system. For a more challenging communication scenario where the LoS path is blocked, the present invention first accurately constructs a near-field channel using the spherical wave hypothesis, and then introduces a designed codebook matrix. By fully exploring the structural relationship between the codebook matrix and the parameters to be estimated, a preliminary extraction of the channel parameters is obtained. Subsequently, the steepest descent method is used to iteratively optimize the parameters with super-resolution. Finally, by analyzing the geometric constraints of the system and combining them with the obtained high-precision parameters, accurate positioning of the user equipment and scatterers is achieved. The proposed method is superior to existing competing methods in terms of parameter estimation accuracy and positioning accuracy, and can still maintain excellent performance under the conditions of low sampling numbers and low pilot overhead.

[0109] As described above, it is only one of the specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be thought of by those skilled in the art within the technical scope disclosed by the present invention without creative labor should be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope defined by the claims.

Claims

1. An effective method for near-field parameter estimation and positioning in a very large-scale multiple-input multiple-output XL-MIMO system, characterized in that The method includes: In a near - field communication scenario where the line - of - sight (LoS) path is blocked, for a system with ultra - large - scale uniform linear arrays deployed at both the transmitter and receiver ends, a spherical - wave hypothesis is adopted to construct a near - field channel and received - signal model; Design a sampling codebook matrix corresponding to angle and distance parameters. According to the correlation between the received signal and the codebook matrix, use the idea of the orthogonal matching pursuit algorithm to solve for the index information related to the near - field parameters; Analyze the association between the obtained index and the parameters to be estimated. According to the structural characteristics of the codebook matrix and the Kronecker inner - product operation rule, realize the preliminary extraction of the angle parameters and distance parameters; Design an optimization function to perform super - resolution refinement on the obtained parameters, and use the idea of the steepest descent to iteratively update the angle and distance parameters; Use the geometric constraints of the system and the estimated angle and distance parameters to sense the key nodes of the system, thereby realizing the accurate positioning of the user equipment and environmental scatterers; Design a sampling codebook matrix corresponding to the design angles and distance parameters. According to the correlation between the received signal and the codebook matrix, use the idea of the orthogonal matching pursuit algorithm to solve the index information related to the near-field parameters, specifically including: through uniform sampling, obtain the sampling values of the angles and distances at the transceiver ends. The angle sampling values corresponding to the transmitter end. The angle sampling values corresponding to the receiver end. The distance sampling values corresponding to the transmitter end, and The distance sampling values corresponding to the receiver end, where K is the number of angle samplings at the transmitter end, T is the number of angle samplings at the receiver end, J is the number of distance samplings at the transmitter end, and U is the number of distance samplings at the receiver end. For the convenience of subsequent parameter extraction, here define θ = [θ1,..., θ K , φ = [φ1,..., φ T , γ = [γ1,..., γ J , and ν = [ν1,..., ν U . Substitute the above parameter sampling information into the steering vector expression corresponding to the spherical wave model and combine them to design the codebook matrices B T and B R , written as Among them, a T (·) is the steering vector at the transmitting end, a R (·) is the steering vector at the receiving end, N T is the number of transmitting antennas, S T =K×J, N R is the number of receiving antennas, S R =T×U, given the initial residual R = Y and the initial sensing matrix Y is the received signal, the vector w = 0 1×1 is the vector that stores the indices obtained in each iteration. The number of iterations is equal to the number of paths L. By virtue of the correlation of the matrix structure, the corresponding index information can be obtained through the idea of greedy search. Define Ω R =Q H B R and (·) H represents the conjugate transpose operation. The residual is sparsified to obtain the sparse representation of the residual matrix in the angular domain and the range domain as follows Vectorize Υ, solve for the index x corresponding to the maximum value in the obtained vector, and accordingly update w = [w, x]. Since x carries information related to the target angle and distance parameters, by analyzing the internal structural association between the codebook matrix and the channel matrix, the update of the sensing matrix can be realized. The specific operations are as follows Among them, x R = x - (x T - 1)S R , (·) T represents the transpose operation, the symbol represents the Kronecker product, represents the ceiling operation, and the sparse channel is reconstructed by Γ to obtain Among them, (·) -1 represents the inverse operation, and vec(·) represents the vectorization operation, mapping the obtained vector into a matrix form, that is Immediately update the residual as follows And repeat the above process until the maximum number of iterations is reached; Design an optimization function to perform super - resolution refinement on the obtained parameters, and use the idea of the steepest descent to iteratively update the angle and distance parameters, specifically including: design the optimization function as follows where P is the precoding matrix, is the estimated channel matrix, Q is the combining matrix, ||·|| F denotes the operation of taking the Frobenius norm, κ is the given regularization parameter, is a diagonal matrix, |·| is the operation of taking the modulus, ο is a constant to ensure the logarithm function is meaningful, is the estimated vector of the angle parameter at the transmitter end, is the estimated vector of the distance parameter at the transmitter end, is the estimated vector of the angle parameter at the receiver end, is the estimated vector of the distance parameter at the receiver end. Specifically, given the number of iterations N iter , the i-th (i = 0, …, N iter - 1) iteration is as follows. Take the partial derivative of the optimization function F with respect to to obtain the optimal solution of the path parameter, and substitute the expression of into the optimization function to obtain Among them, Next, perform the following parameter update where η is the update step size, is the partial derivative of F (i) with respect to corresponding to the gradient descent direction of the transmitter angle parameter, is the partial derivative of F (i) with respect to corresponding to the gradient descent direction of the transmitter distance parameter, is the partial derivative of F (i) with respect to corresponding to the gradient descent direction of the receiver angle parameter, is the partial derivative of F (i) with respect to corresponding to the gradient descent direction of the receiver distance parameter. After completing the update of using the steepest descent idea, the corresponding is obtained. Repeat this process until the iteration ends, and denote the final result as Substitute the obtained parameters into the channel model, and the final estimate of the near-field channel can be obtained.

2. An effective near-field parameter estimation and positioning method for an XL-MIMO system according to claim 1, characterized in that In a near-field communication scenario where the LoS path is blocked, for a system with ultra-large-scale uniform linear arrays deployed at both the transmitter and receiver ends, a spherical wave assumption is adopted to construct the near-field channel and received signal models, specifically including: considering a communication system operating in the millimeter-wave band, where both the user equipment and scatterers are located within the near-field radiation region. In this system, both the base station side and the user equipment side are equipped with ultra-large-scale uniform linear arrays, with the number of antennas being N T and N R respectively. The LoS path between the base station and the user equipment is blocked, and the system relies solely on the non-line-of-sight (NLoS) path for communication. Considering the electromagnetic wave characteristics of the near field, a spherical wave model is used for channel modeling, and the expression of the corresponding channel is written as where L is the number of paths, and z l is the path gain corresponding to the l-th path, l = 1, …, L, θ l and correspond to the departure angle and arrival angle in the l-th path respectively, d T,l is the distance between the center point of the transmitting array and the l-th scatterer, d R,l is the distance between the center point of the receiving array and the l-th scatterer, a T (·) is the steering vector of the transmitting end, a R (·) is the steering vector of the receiving end, and their expressions are as follows where f is the carrier frequency, c is the speed of light, and is the distance between the n T (n T = 1, …, N T )th antenna of the base station antenna array and the lth scatterer, is the distance between the n R (n R = 1, …, N R )th antenna of the user equipment antenna array and the lth scatterer, and is expressed as Among them, For the convenience of subsequent processing, the matrix mode of the channel H is also given accordingly, that is Among them, A T = [a T (θ1, d T,1 ), …, a T (θ L , d T,L )], z = [z1, …, z L , assuming that the base station sends N X pilot signals, then the signal received at the user equipment side is written as Y = Q H HP + N, Among them, is a precoding matrix that satisfies the constant modulus constraint, is a combining matrix that satisfies the constant modulus constraint, N Y is the corresponding received signal dimension, is a noise matrix that follows a Gaussian distribution.

3. An effective near-field parameter estimation and positioning method for an XL-MIMO system according to claim 1, characterized in that Analyze the correlation between the obtained index and the parameter to be estimated, and based on the structural characteristics of the codebook matrix and the Kronecker inner product algorithm, realize the preliminary extraction of the angle parameter and the distance parameter, specifically including: Denote As can be seen from the analysis, each row in χ corresponds to a set of sampling values of the angle and distance parameters. Therefore, using the obtained index information and the algorithm of the Kronecker product, realize the preliminary extraction of the angle and distance parameters, so as to obtain where Correspondingly, the path gain is solved by the idea of least squares, and its solution process is as follows where C = [c1, …, c L , So far, the method has obtained a preliminary estimate including the departure angle, arrival angle, distance between the center of gravity of the base station array and the scattering point, distance between the center of gravity of the user equipment array and the scattering point, and path gain, that is 4. An effective near-field parameter estimation and positioning method for an XL-MIMO system according to claim 1, characterized in that Use the geometric constraints of the system and the estimated angle and distance parameters to sense the key nodes of the system, thereby realizing the accurate positioning of the user equipment and environmental scatterers, specifically including: perform geometric analysis on the system, combine the obtained high - precision parameters with the geometric relationship between nodes, and realize the positioning of the scatterers in the environment. Specifically, the position of the scatterer corresponding to the l - th path will be calculated by the following formula, that is, After obtaining the position information of the scatterers corresponding to all paths, regard the obtained scatterer position information as known information, thereby assisting in the positioning of the user equipment. The formula for solving the position of the user equipment is given as follows Thus, the accurate estimation of the positions of the user equipment and scatterers in the near - field environment is realized.

Images