A Smart Vehicle Localization Method Based on Enhanced DOA Estimation Using 5G Cooperative Base Stations

Abstract

This invention relates to a method for intelligent vehicle positioning based on enhanced DOA estimation using 5G base stations. The method includes receiving intelligent vehicle positioning signals using large-scale uniform linear arrays from three cooperating 5G base stations; determining the DOA estimation signal form for each base station antenna array after considering the unknown mutual coupling effects of the arrays; calculating the covariance matrix of the observation data from each array; sequentially applying mutual coupling linear transformation suppression, Toeplitz integration, and signal subspace phase compensation; obtaining enhanced DOA estimation based on subarray partitioning technology; optimizing all DOAs based on the DOA estimation results obtained from the three cooperating 5G base stations using the received signal-to-noise ratio information and DOA size obtained from each base station; and finally obtaining the intelligent vehicle's location information based on triangulation using the optimized DOAs. Compared with existing methods, this invention not only has lower complexity but also provides higher resolution, positioning accuracy, and positioning robustness.

Description

Technical Field

[0001] This invention relates to the fields of vehicle networking and intelligent transportation, and in particular to an intelligent vehicle localization method based on enhanced DOA estimation using 5G cooperative base stations. Background Technology

[0002] Intelligent vehicles represent a breakthrough point for the transformation and upgrading of the automotive industry in the new era and a strategic high ground for the future. They are not merely technological advancements in automobiles themselves, but also products of the integration of new technologies such as the Internet of Things (IoT), cloud computing, big data, and mobile internet. Intelligent vehicles, capable of comprehensively achieving safety, energy conservation, environmental protection, comfortable driving, and improved traffic efficiency, have become a hot topic and development direction in the fields of vehicle networking and intelligent transportation. Intelligent vehicles are a multi-party collaborative ecosystem based on autonomous driving and vehicle networking technologies. In this system, high-precision vehicle positioning technology is one of the key technologies and a crucial guarantee for safe vehicle passage. In most vehicle networking application scenarios, vehicle positioning based on the Global Navigation Satellite System (GNSS) is the most common and basic solution. However, considering factors such as environment (obstruction, lighting, weather), positioning accuracy, cost, and stability, simply using GNSS positioning technology cannot meet the positioning requirements of vehicle networking and intelligent transportation services. Against this backdrop, improving vehicle positioning accuracy, continuity, and stability through other methods has become a core development trend in vehicle networking.

[0003] According to the "White Paper on High-Precision Vehicle Positioning" released by the IMT-2020 (5G) Promotion Group and the latest 5G wireless positioning literature, relying on 5G wireless communication equipment (such as various wireless access devices and cellular cooperative base stations) or various advanced sensors (such as sonar, lidar, and cameras) combined with mechanisms such as Time of Arrival (TOA), Time Difference of Arrival (TDOA), and Direction of Arrival (DOA) estimation has become the mainstream solution for intelligent vehicle positioning. Among these solutions, positioning technology based on DOA estimation of 5G cooperative base station antennas has become one of the positioning technologies with wide application value because it can complete all-weather positioning of multiple vehicles and is relatively insensitive to the accuracy of time delay measurement. It plays a crucial role in improving the accuracy and robustness of vehicle positioning and can effectively compensate for the shortcomings of GNSS positioning in various application scenarios.

[0004] In recent years, two representative schemes have emerged for intelligent vehicle localization based on DOA estimation using cooperative base stations. The first scheme was proposed by H. Wang et al. in 2019 in the paper "Assistant vehicle localization based on three collaborative base stations via SBL-based robust DOAestimation". This scheme achieves DOA estimation and vehicle localization based on sparse Bayesian techniques under three cooperative base stations. The second scheme was proposed by F. Wen et al. in 2020 in the paper "Auxiliary vehicle positioning based on robust DOAestimation with unknown mutual coupling". This scheme considers the influence of actual array antenna mutual coupling and obtains DOA estimation and vehicle localization based on the rank deficiency criterion. Both of the above schemes are beneficial attempts and explorations of intelligent vehicle positioning technology based on cooperative base station DOA estimation. However, analysis revealed that the first scheme did not consider the actual array mutual coupling effect and had high complexity, making it difficult to apply in practice. The second scheme is based on the classical asymptotic system, which has a significant threshold collapse phenomenon. Furthermore, when applied under 5G large-scale arrays, the vehicle positioning robustness is difficult to guarantee due to the inconsistent estimation of the sample covariance matrix, eigenvalues ​​and eigenvectors. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide an intelligent vehicle positioning method based on 5G cooperative base station enhanced DOA estimation with low computational complexity and guaranteed vehicle positioning robustness when applied in large-scale arrays.

[0006] The technical solution adopted in this invention is a smart vehicle positioning method based on 5G cooperative base station enhanced DOA estimation, which includes the following steps:

[0007] Step 1: Utilize the large-scale uniform linear arrays in three cooperating 5G base stations to receive the positioning signal of the intelligent vehicle and determine the DOA estimation signal form under the antenna array of each cooperating base station.

[0008] Step 2: Calculate the covariance matrix of the observation data of the antenna arrays of each cooperating base station. Its expression is: Where N represents the number of sampling times, y i (t) represents the array output data of the i-th base station at time t. Each covariance matrix is ​​then subjected to mutual coupling linear transformation suppression, Toeplitz integration, and signal subspace phase compensation in sequence. Then, based on the subarray partitioning technique, the enhanced DOA estimation results of each cooperative 5G base station are obtained.

[0009] Step 3: Based on the DOA estimation results of the three cooperative 5G cooperative base stations obtained in Step 2, use the received signal-to-noise ratio information obtained by each cooperative base station to optimize all the DOA estimation results to obtain the optimized DOA estimation results. Finally, according to the optimized DOA estimation results, use the triangulation method to obtain the position information of the intelligent vehicle.

[0010] Preferably, in Step 1, the specific process of determining the DOA estimation signal form under the antenna array of each cooperative base station is as follows: Assume that the uncorrelated positioning signals transmitted by K intelligent vehicles are incident on the uniform linear array of the i-th 5G cooperative base station, where i = 1, 2, 3; the number of array elements is M, and the element spacing is d. Considering the unknown mutual coupling of the array, the DOA estimation signal form of the array corresponding to the i-th 5G cooperative base station at time t is expressed as:

[0011] Among them, s(t) = [s1(t),..., s K (t)] T , represents the output data of the M-th antenna in the i-th base station at time t, s K (t) represents the value of the K-th signal at time t, represents the noise data of the M-th antenna in the i-th base station at time t represents the steering vector of the i-th base station receiving the k-th signal, represents the steering value corresponding to the k-th signal of the M-th antenna, λ is the carrier wavelength, λ ≥ 2d, is the complex-valued mutual coupling coefficient, satisfying and P < M, t represents the sampling time, the superscript T represents the transpose operation, Toeplitz{r} represents the symmetric Toeplitz matrix composed of the elements of the vector r, k = 1,..., K;

[0012] Preferably, in Step 2, the specific process of obtaining the enhanced DOA estimation results of each cooperative 5G cooperative base station by sequentially passing each covariance matrix through mutual coupling linear transformation suppression, Toeplitz integration, and signal subspace phase compensation, and then based on the subarray partitioning technique includes the following steps:

[0013] (2.1) Based on the symmetric Toeplitz property of the mutual coupling matrix, use the linear transformation matrix to suppress the mutual coupling influence between the linear arrays. The expression of the linear transformation matrix is: Among them, T = [0 (M-2P)×P , IM-2P ,0 (M-2P)×P ], 0 (M-2P)×P Let I represent an (M-2P)×P dimensional all-zero matrix. M-2P Denotes an (M-2P)×(M-2P) dimensional identity matrix. Indicates that by A i The middle of A matrix consisting of rows of elements, P c It is a diagonal matrix containing the positioning signal power and the mutual coupling coefficient. Noise power;

[0014] (2.2) The result obtained in step (2.1) Toeplitz integration yields the enhanced covariance matrix Ξ i Its expression is: Among them, J m express Transformation matrix, The m-th diagonal element of the transformation matrix is ​​1, and all other elements are 0. -m Represented as (J) m ) T ,and

[0015] (2.3) The enhanced covariance matrix Ξ in step (2.2) i Perform eigenvalue decomposition to obtain the signal subspace. and their corresponding eigenvalues After performing phase compensation on the obtained signal subspace, the improved signal subspace matrix is ​​obtained, and its expression is: in,

[0016] (2.4) Using subarray partitioning techniques, the subarrays in step (2.3) are partitioned. Divided into and express The former OK, express After Okay, according to and get: For Γ i Perform eigenvalue decomposition to obtain K eigenvalues.

[0017] (2.5) Based on the K feature values ​​obtained in step (2.4), the estimated DOA of the positioning signal transmitted by the k-th intelligent vehicle reaching the i-th cooperating base station is: in, Represents retrieving elements Complex angle operations.

[0018] Preferably, in step 3, the specific process of using the received signal-to-noise ratio information obtained from each cooperating base station to optimize all DOA estimation results, obtaining optimized DOA estimation results, and finally using triangulation to obtain the location information of the intelligent vehicle based on the optimized DOA estimation results includes the following steps:

[0019] (3.1) The received signal-to-noise ratios (SNRs) for the same vehicle obtained by the three cooperative 5G base stations are sorted in descending order, and their received SNRs are denoted as SNR1, SNR2 and SNR3 respectively.

[0020] (3.2) If SNR2-SNR3>0.5 dB, then the DOA estimation results corresponding to SNR1 and SNR2 are selected as the preferred DOA estimation results; if SNR2-SNR3≤0.5 dB, then the magnitude of the DOA estimation result corresponding to SNR2 is compared with the magnitude of the DOA estimation result corresponding to SNR3, and the DOA estimation result with the smallest magnitude and the DOA estimation result corresponding to SNR1 are selected as the preferred DOA estimation results.

[0021] (3.3) Based on the optimized DOA estimation results, the location information of the intelligent vehicle is obtained by triangulation.

[0022] In step (3.3), based on the preferred DOA estimation result, the specific process of obtaining the intelligent vehicle's location information using triangulation is as follows: From the three received signal-to-noise ratios (SNRs), select the two SNRs with the larger SNR values. The location information of the cooperating base stations corresponding to these two SNRs is represented as (L... 1x ,L 1y ) and (L 2x ,L 2y The DOA estimation results corresponding to the two received signal-to-noise ratios are respectively expressed as θ. k1 and θ k2 The location information of the intelligent vehicle is then represented as:

[0023]

[0024]

[0025] Compared with existing technologies, the beneficial effects of this invention are as follows: The method of this invention obtains enhanced DOA estimation results based on Toeplitz integration and signal subspace phase compensation techniques under the theory of large-dimensional random matrices. This effectively avoids the non-uniform estimation problem existing in the application of classical asymptotic system methods under 5G large-scale arrays, effectively improving the accuracy of DOA estimation and associated intelligent vehicle positioning. Furthermore, the method of this invention is applicable to unknown array coupling conditions and obtains low-complexity DOA estimation through closed-form solutions, effectively ensuring the real-time requirements of intelligent vehicle positioning. After obtaining the DOA estimation results provided by three cooperative base stations, the method provided by this invention selects the optimal DOA estimation result through auxiliary information before performing vehicle positioning, effectively avoiding the impact of unstable DOA on vehicle positioning performance. Attached Figure Description

[0026] Figure 1 This is a schematic diagram of intelligent vehicle positioning based on enhanced DOA estimation using three 5G cooperative base stations in this invention;

[0027] Figure 2 This is a flowchart of an intelligent vehicle localization method based on enhanced DOA estimation using 5G cooperative base stations, as proposed in this invention.

[0028] Figure 3 This is a comparison chart showing the absolute positioning error of the intelligent vehicle positioning method based on 5G cooperative base station enhanced DOA estimation as a function of signal-to-noise ratio in a specific embodiment of the present invention.

[0029] Figure 4 This is a comparison chart showing the variation of the single-run average time of the intelligent vehicle positioning method based on 5G cooperative base station enhanced DOA estimation proposed in this invention with the number of array elements and the number of sampling snapshots. Detailed Implementation

[0030] The invention will be further described below with reference to the accompanying drawings and specific embodiments, so that those skilled in the art can implement it based on the description. The scope of protection of the invention is not limited to these specific embodiments.

[0031] This invention provides a method for intelligent vehicle localization based on enhanced DOA estimation using 5G cooperative base stations, the implementation steps of which are as follows:

[0032] Step 1: Utilize the large-scale uniform linear arrays in three cooperating 5G base stations to receive the positioning signal of the intelligent vehicle. After considering the unknown mutual coupling effect of the arrays, determine the DOA estimation signal form under the antenna array of each cooperating base station.

[0033] In this embodiment, the specific process of determining the DOA estimation signal form under each cooperative base station antenna array is as follows: Assume that the uncorrelated positioning signals transmitted by K intelligent vehicles are incident on the uniform linear array of the i-th (i = 1, 2, 3) 5G cooperative base station, the number of array elements is M, and the element spacing is d. Considering the unknown mutual coupling of the array, the DOA estimation signal form of the array corresponding to the i-th 5G cooperative base station at time t is expressed as:

[0034]

[0035] Where s(t) = [s1(t),..., s K (t)] T , represents the output data of the M-th antenna in the i-th base station at time t, s K (t) represents the value of the K-th signal at time t, represents the noise data of the M-th antenna in the i-th base station at time t represents the steering vector of the i-th base station receiving the k-th signal, represents the steering value of the M-th antenna corresponding to the k-th signal, λ is the carrier wavelength, λ ≥ 2d, is the complex-valued mutual coupling coefficient, satisfying and P < M, t represents the sampling time, the superscript T represents the transpose operation, Toeplitz{r} represents the symmetric Toeplitz matrix composed of the elements of the vector r, k = 1,..., K;

[0036] Step 2: Calculate the covariance matrix of the observation data of the linear array of each 5G cooperative base station, and its expression is: Where, N represents the number of sampling times, y i (t) represents the array output data of the i-th base station at the t-th time. The covariance matrix of the observation data of the linear array of each obtained 5G cooperative base station is successively subjected to mutual coupling linear transformation suppression, Toeplitz integration, and signal subspace phase compensation, and then an enhanced DOA estimation result corresponding to each 5G cooperative base station is obtained based on the subarray division technology;

[0037] In this embodiment, after successively performing mutual coupling linear transformation suppression, Toeplitz integration, and signal subspace phase compensation, the specific process of obtaining the enhanced DOA estimation result based on the subarray division technology includes the following steps:

[0038] (2.1), Based on the symmetric Toeplitz property of the mutual coupling matrix, use the linear transformation matrix to suppress the mutual coupling influence between linear arrays. The expression of the linear transformation matrix is: Where, T = [0 (M-2P)×P ,I M-2P ,0 (M-2P)×P ], 0 (M-2P)×P Let I be an (M-2P)×P dimensional all-zero matrix. M-2P It is an (M-2P)×(M-2P) dimensional identity matrix. For A i The middle A matrix consisting of rows of elements, P c It is a diagonal matrix containing the positioning signal power and the mutual coupling coefficient. Noise power;

[0039] (2.2) The result obtained in step (2.1) Toeplitz integration yields the enhanced covariance matrix Ξ i :

[0040]

[0041] Among them, J m for The transformation matrix has its m-th top diagonal element set to 1, and all other elements set to 0. J -m Representative (J) m ) T ,and

[0042] (2.3) The enhanced covariance matrix Ξ obtained in step (2.2) i Perform eigenvalue decomposition to obtain the signal subspace. and the corresponding eigenvalues By utilizing the limiting property of eigenvectors in the theory of large-dimensional random matrices, phase compensation is performed on the signal subspace to obtain an improved signal subspace matrix, the expression of which is:

[0043]

[0044] in, c = M / N;

[0045] (2.4) Using subarray partitioning techniques to divide Divided into and and They represent The former lines and After Okay, and according to and Further obtain For Γ i Eigenvalue decomposition yields K eigenvalues

[0046] (2.5) Based on step (2.4), K feature values ​​are obtained to obtain the estimated DOA value of the positioning signal transmitted by the k-th intelligent vehicle reaching the i-th cooperating base station:

[0047]

[0048] in, Represents retrieving elements Complex angle operations.

[0049] Step 3: Based on the DOA estimation results of the three 5G cooperative base stations obtained in Step 2, the received signal-to-noise ratio information obtained by each cooperative base station is used to optimize all the DOA estimation results to obtain the optimized DOA estimation results. Finally, based on the optimized DOA estimation results, the triangulation method is used to obtain the location information of the intelligent vehicle.

[0050] In this embodiment, the process of optimizing all DOA estimation results by utilizing the received signal-to-noise ratio information obtained from each cooperating base station to obtain an optimized DOA estimation result, and finally obtaining the location information of the intelligent vehicle using triangulation based on the optimized DOA estimation result, includes the following steps:

[0051] (3.1) The received signal-to-noise ratios obtained by the three 5G cooperative base stations for the same vehicle are sorted in descending order and denoted as SNR1, SNR2 and SNR3 respectively; SNR1, SNR2 and SNR3 are the received signal-to-noise ratios after descending order.

[0052] (3.2) If SNR2-SNR3>0.5 dB, then the DOA estimation results corresponding to SNR1 and SNR2 are selected as the preferred DOA estimation results; if SNR2-SNR3≤0.5 dB, then the DOA estimation result with the smallest modulus and the DOA estimation result corresponding to SNR1 are selected as the preferred DOA estimation results.

[0053] (3.3) Based on the optimized DOA estimation results, the location information of the intelligent vehicle is obtained using triangulation. That is, the two receiving signal-to-noise ratios with larger values ​​are selected from the three receiving signal-to-noise ratios. The location information of the cooperative base stations corresponding to these two receiving signal-to-noise ratios are represented as (L... 1x ,L 1y ) and (L 2x ,L 2y The DOA estimation results corresponding to the two received signal-to-noise ratios are respectively expressed as θ. k1 and θ k2 The location information of the intelligent vehicle is then represented as:

[0054]

[0055]

[0056] The following simulation experiments analyze the positioning performance and computational effectiveness of the intelligent vehicle positioning method based on 5G cooperative base station enhanced DOA estimation proposed in this invention. The simulation process is carried out using MATLAB software, and the computer configuration is a Thinkpad X1 laptop with an i5-10210U CPU and 8G DDR4 RAM.

[0057] Simulation Experiment 1: Using the method proposed in this embodiment of the invention, the rank-deficient method in the prior art, and the ESPRIT method under the classical asymptotic system in the prior art, the relationship between the absolute positioning error (AE, unit: meters (m)) and the signal-to-noise ratio is simulated respectively. The positions of the three 5G cooperative base stations are B1(0m, 500m), B2(0m, 0m), and B3(600m, 0m), respectively. The number of array elements in each cooperative base station is the same and is 50. The number of sampling snapshots is 100. The coordinates of the four intelligent vehicles are located at S1(395.4525m, 356.0671m), S2(370.4931m, 357.7810m), S3(395.7028m, 332.0341m) and S4(85.5050m, 484.9232m), respectively. The mutual coupling matrix is ​​set to... and Consider the following two different scenarios: Scenario (1) The received signal-to-noise ratio (SNR) of each cooperating base station is the same; Scenario (2) The received SNR of each cooperating base station is different, with the SNR of the third cooperating base station being 3 dB lower than that of the first and second cooperating base stations. The simulation results of the received SNR of the first cooperating base station changing from 0 dB to 30 dB are as follows: Figure 3 As shown, where Figure 3 A corresponds to scenario (1). Figure 3 B corresponds to scenario (2). (By...) Figure 3 As can be seen from A, the positioning performance of the method of the present invention is comprehensively superior to the ESPRIT method under the classical asymptotic system. Furthermore, it can achieve sub-meter positioning accuracy when the signal-to-noise ratio is greater than or equal to 10 dB, and even centimeter-level positioning accuracy when the signal-to-noise ratio reaches 25 dB or higher. This fully demonstrates the effectiveness and superiority of the method of the present invention. On the other hand, from Figure 3 B shows that the method of the present invention can still guarantee excellent positioning performance even when the received signal-to-noise ratios of the three cooperative base stations are different, while the comparison method shows a significant performance decline, which fully proves the stability of the method of the present invention.

[0058] Simulation Experiment 2: Using the method proposed in this invention and the rank-deficient method of the prior art, the relationship between the average single running time of vehicle positioning and the number of cooperative base station array elements and the number of sampling snapshots was simulated. The signal-to-noise ratio was fixed at 10 dB, the spatial grid search range of the rank-deficient method was -90° to 90°, and the search step size was 1°. Figure 4 In A, the number of sampling snapshots is fixed at 100, and the number of array elements changes from 30 to 100. Figure 4 In step B, the number of array elements is fixed at 50, while the number of sampling snapshots varies from 50 to 400. Simulation results show that the running time of the method provided by this invention is significantly lower than that of the rank-deficient method in the prior art, which fully demonstrates the computational efficiency of the method provided by this invention.

Claims

1. A method for intelligent vehicle localization based on enhanced DOA estimation using 5G cooperative base stations, characterized in that: The method includes the following steps: Step 1: Utilize the large-scale uniform linear arrays in three cooperating 5G base stations to receive the positioning signal of the intelligent vehicle and determine the DOA estimation signal form under the antenna array of each cooperating base station. Step 2: Calculate the covariance matrix of the observation data of the antenna arrays of each cooperating base station. Its expression is: Where N represents the number of sampling times, y i (t) represents the array output data of the i-th base station at time t. Each covariance matrix is ​​sequentially suppressed by mutual coupling linear transformation, Toeplitz integration, and signal subspace phase compensation. Then, based on the subarray partitioning technique, the enhanced DOA estimation result of each cooperative 5G base station is obtained. The specific process is as follows: (2.1) Based on the symmetric Toeplitz property of the mutual coupling matrix, a linear transformation matrix is ​​used to suppress the mutual coupling effect between linear arrays. The expression of the linear transformation matrix is ​​as follows: Where, T = [0 (M-2P)×P ,I M-2P ,0 (M-2P)×P ], 0 (M-2P)×P Let I represent an (M-2P)×P dimensional all-zero matrix. M-2P Denotes an (M-2P)×(M-2P) dimensional identity matrix. Indicates that by A i The middle of A matrix consisting of rows of elements, P c It is a diagonal matrix containing the positioning signal power and the mutual coupling coefficient. Noise power; (2.2) The result obtained in step (2.1) Toeplitz integration yields the enhanced covariance matrix Ξ i Its expression is: Among them, J m express Transformation matrix, The m-th diagonal element of the transformation matrix is ​​1, and all other elements are 0. -m Represented as (J) m ) T ,and (2.3) The enhanced covariance matrix Ξ in step (2.2) i Perform eigenvalue decomposition to obtain the signal subspace. and their corresponding eigenvalues After performing phase compensation on the obtained signal subspace, the improved signal subspace matrix is ​​obtained, and its expression is: in, (2.4) Using subarray partitioning techniques, the subarrays in step (2.3) are partitioned. Divided into and express The former OK, express After Okay, according to and get: For Γ i Perform eigenvalue decomposition to obtain K eigenvalues. (2.5) Based on the K feature values ​​obtained in step (2.4), the estimated DOA of the positioning signal transmitted by the k-th intelligent vehicle reaching the i-th cooperating base station is: in, Represents retrieving elements Complex angle operations; Step 3: Based on the DOA estimation results of the three cooperative 5G base stations obtained in Step 2, the received signal-to-noise ratio information obtained by each cooperative base station is used to optimize all the DOA estimation results to obtain the optimized DOA estimation results. Finally, based on the optimized DOA estimation results, the triangulation method is used to obtain the location information of the intelligent vehicle.

2. The intelligent vehicle positioning method based on 5G cooperative base station enhanced DOA estimation according to claim 1, characterized in that: In step 1, the specific process of determining the DOA estimation signal form under the antenna array of each cooperating base station is as follows: Assume that K uncorrelated positioning signals emitted by intelligent vehicles are incident on the uniform linear array of the i-th 5G cooperating base station, where i = 1, 2, 3; the number of array elements is M, and the element spacing is d. Considering the unknown mutual coupling of the array, the DOA estimation signal form of the array corresponding to the i-th 5G cooperating base station at time t is expressed as: Among them, s(t) = [s1(t),..., s K (t)] T , denotes the output data of the M-th antenna in the i-th base station at time t, s K (t) represents the value of the K-th signal at time t, denotes the noise data of the M-th antenna in the i-th base station at time t denotes the steering vector for the i-th base station to receive the k-th signal, denotes the steering value corresponding to the k-th signal of the M-th antenna, λ is the carrier wavelength, λ ≥ 2d, is the complex-valued mutual coupling coefficient, satisfying and P < M, t represents the sampling time, the superscript T represents the transpose operation, Toeplitz{r} represents the symmetric Toeplitz matrix composed of the elements of the vector r, k = 1,..., K.

3. The intelligent vehicle positioning method based on 5G cooperative base station enhanced DOA estimation according to claim 2, characterized in that: In step 3, the specific process of using the received signal-to-noise ratio information obtained from each cooperating base station to optimize all DOA estimation results, obtaining optimized DOA estimation results, and finally using triangulation to obtain the location information of the intelligent vehicle based on the optimized DOA estimation results includes the following steps: (3.1) The received signal-to-noise ratios (SNRs) for the same vehicle obtained by the three cooperative 5G base stations are sorted in descending order, and their received SNRs are denoted as SNR1, SNR2 and SNR3 respectively. (3.2) If SNR2-SNR3>0.5 dB, then the DOA estimation results corresponding to SNR1 and SNR2 are selected as the preferred DOA estimation results; if SNR2-SNR3≤0.5 dB, then the magnitude of the DOA estimation result corresponding to SNR2 is compared with the magnitude of the DOA estimation result corresponding to SNR3, and the DOA estimation result with the smallest magnitude and the DOA estimation result corresponding to SNR1 are selected as the preferred DOA estimation results. (3.3) Based on the optimized DOA estimation results, the location information of the intelligent vehicle is obtained by triangulation. In step (3.3), based on the preferred DOA estimation result, the specific process of obtaining the intelligent vehicle's location information using triangulation is as follows: From the three received signal-to-noise ratios (SNRs), select the two SNRs with the larger SNR values. The location information of the cooperating base stations corresponding to these two SNRs is represented as (L... 1x ,L 1y ) and (L 2x ,L 2y The DOA estimation results corresponding to the two received signal-to-noise ratios are respectively expressed as θ. k1 and θ k2 The location information of the intelligent vehicle is then represented as:

Images