Sensitivity integrated unmanned aerial vehicle positioning method based on multi-base-station weighted fusion

Abstract

The invention belongs to the field of wireless communication, and particularly relates to a multi-base-station weighted fusion-based communication and sensing integrated unmanned aerial vehicle positioning method. According to the invention, a weighted accumulation method is utilized to fuse unmanned aerial vehicle coordinates estimated by each base station so as to complete positioning. Specifically, a plurality of communication sensing integrated base stations are utilized, each base station sends a communication signal based on an OFDM waveform to the unmanned aerial vehicle, and the relative position of the unmanned aerial vehicle in each base station is detected by processing the symbol difference between a transmitted signal and a received signal. A plurality of absolute coordinates of the unmanned aerial vehicle can be estimated according to the known and accurate absolute coordinates of the base stations. On the basis, estimated values of a plurality of absolute coordinates are synthesized into a central coordinate according to a signal-to-noise ratio weighting mode, and a three-dimensional grid is constructed by taking the coordinate as a center. And a new spectral function is constructed according to the signal-to-noise ratio of the channel, the function value of each point on the grid is calculated, and the point corresponding to the maximum value is the final positioning coordinate of the unmanned aerial vehicle.

Description

Technical Field

[0001] The present invention belongs to the field of Integrated Sensing And Communications (ISAC) in wireless communications, and specifically relates to a synaesthesia integrated unmanned aerial vehicle positioning method based on weighted fusion of multiple base stations. Background Art

[0002] As the main carrier of low-altitude activities, UAV (Unmanned Aerial Vehicle) has unique advantages such as high flexibility, good controllability and rapid deployment. Its typical application scenarios include express logistics, geographic surveying and mapping, urban management, emergency rescue, energy inspection, traffic inspection, agricultural and forestry plant protection, etc., and it has broad market prospects.

[0003] In practical applications, it is crucial to determine the exact location of drones. However, when drones fly in complex environments, especially high-rise buildings in cities, the positioning performance of the Global Navigation Satellite System (GNSS) is greatly limited due to the obstruction of obstacles. On the other hand, the Visual Positioning System (VPS) is sensitive to lighting and occlusion and is constrained by environmental conditions. The positioning stability of drones is relatively poor, and new technologies are urgently needed.

[0004] As an emerging technology of the sixth generation mobile communication (6G), the communication and perception integration technology has attracted widespread attention in academia and industry. The integrated design of communication and perception can enable these two originally independent applications to be implemented in the same system and benefit each other. The perception capability of the integrated system of communication aims to realize the active or passive perception function of the UAV by using wireless signals. By effectively extracting the wireless signal characteristics of the target, the functions of ranging, angle measurement, speed measurement, positioning, and tracking of the target can be realized. This information can help enhance the safety of flight missions and improve their operational efficiency. In addition, the communication and perception integration system can replace independent communication and perception systems, reduce the flight load of the UAV, thereby improving the flexibility of the UAV, and then improve its flight time, flight distance and actual load, which has great potential.

[0005] Compared with traditional positioning technology, the integrated communication and perception system perceives the characteristics of the signal system, has a certain anti-interference ability, can adjust the detection parameters in real time to adapt to environmental changes, and improve the accuracy of positioning, especially in complex urban environments.

[0006] However, the traditional single-base station positioning algorithm currently used in the integrated communication and perception system still has insufficient positioning accuracy in practical applications. Especially when the channel signal-to-noise ratio is low, the positioning error exceeds the meter level and cannot be applied to actual scenarios. The improved multi-base station average positioning algorithm ignores the differences in the communication channels between each base station and the drone, which will amplify the results of the poor channel, resulting in poor positioning error stability. There is a high probability that the error of the multi-base station average positioning is higher than the error of the single base station positioning. Summary of the invention

[0007] The purpose of the present invention is to propose a synaesthesia integrated UAV positioning method based on multi-base station weighted fusion in view of the above limitations, so as to improve the positioning accuracy of the UAV coordinates in low-altitude scenes.

[0008] The technical solution of the present invention is: using a weighted accumulation method to fuse the coordinates of the drone estimated by each base station to complete the positioning. The definition of the synaesthesia integrated positioning system includes 1 drone and N communication perception integrated base stations. Each base station sends a communication signal based on an OFDM waveform to the drone, and uses an antenna array to receive the reflected signal from the drone. By processing the symbol difference between the transmitted signal and the received signal, the multiple signal classification (MUSIC) algorithm is used to detect the relative position of the drone to each base station. The estimated value of the absolute coordinates of the drone can be calculated from the known and accurate absolute coordinates of each base station. On this basis, the estimated values ​​of multiple absolute coordinates are synthesized into a central coordinate in a signal-to-noise ratio weighted manner, and a three-dimensional grid is constructed with the coordinate as the center. A new spectral function is constructed according to the signal-to-noise ratio of the channel, and then the function value of each point on the grid is calculated. The point corresponding to the maximum value is the final positioning coordinate of the drone. The feature of the present invention includes the following steps:

[0009] S1. Constructing a synaesthesia integrated positioning system: N communication and perception integrated base stations are placed on the ground, each of which is equipped with a planar antenna array facing upward, and a drone flies above each base station. First, the real coordinates of the drone and the synaesthesia integrated base station are initialized. Each base station transmits an OFDM-based communication and perception integrated waveform to the drone and receives the reflected wave from the drone for processing.

[0010] S2. Each base station processes its transmit and receive signals respectively, and detects the incident angle and relative distance by comparing the changes in the transmit and receive signals to obtain the relative position of the drone. Dividing the transmit signal matrix by the receive signal matrix, the expression of the response of the OFDM waveform corresponding to the μth symbol on the mth subcarrier of the nth carrier is as follows:

[0011]

[0012] Where R n With v n is the distance and radial velocity of the UAV in the direction corresponding to the nth base station, U n is the amplitude attenuation, η is the noise, the noise follows a Gaussian distribution with a mean of 0 and a variance of σ 2 .

[0013] First, the elevation and azimuth angles are estimated. Each base station is equipped with a planar antenna array of size P×Q, placed parallel to the XOY plane. n and Respectively represent the azimuth and elevation of the received signal. If the drone is above the base station, the pitch angle The value range is (0,π / 2), the azimuth angle θ n The value range is (-π,π). Taking the first unit on the upper left of the antenna array as the reference unit, detect the phase difference caused by the azimuth and elevation angles on different elements of the antenna array. The phase difference between the antenna unit in the pth row and the qth column and the reference unit can be expressed as a steering vector, which is in

[0014]

[0015] The signals received by different base stations can be expressed as

[0016] Y r,n =a n s r,n +N n (Formula 4)

[0017] where the vector s r,n is the nth base station echo signal matrix C n The rth row (column), N n represents additive white Gaussian noise.

[0018] Receive angle signal matrix Y a,r,n The size is PQ×N s , and its autocorrelation matrix is ​​calculated as

[0019]

[0020] R a Perform eigenvalue decomposition and construct the noise subspace U a,r,n =U x (:,2:P×Q), where U x For R aThe characteristic matrix is ​​arranged in descending order according to the size of the eigenvalues. The estimated values ​​of the elevation angle and the azimuth angle can be obtained by searching for the maximum value of the spectrum function. Assuming that the communication channel is stable during the sampling time and the frequency response within the bandwidth is approximately the same, the estimation accuracy can be improved by superimposing the spectrum functions of multiple subcarriers. On this basis, the following formula can be obtained. Then, finding the peak value can reduce the influence of random noise. The high-precision estimation results of the azimuth angle and the elevation angle are recorded as and

[0021]

[0022] Then the distance estimation is carried out. Similar to the process of angle estimation, the characteristic matrix U of the distance autocorrelation matrix is ​​first calculated. d,c,n , and then use the distance spectrum function superposition of multiple subcarriers to obtain a high-precision estimate, that is,

[0023]

[0024] By searching for the above peaks, we can get high-precision estimates of distance and speed.

[0025] For positioning, given and The relative coordinates can be obtained. By knowing the position of the corresponding base station, each single base station can calculate a set of absolute coordinates of the drone, which can be recorded as

[0026] S3, according to the measured signal-to-noise ratio between the UAV and each base station, a new absolute coordinate is synthesized, that is,

[0027]

[0028] S4, with new coordinates As the center, construct an n×n×n three-dimensional grid, and the coordinate spacing between adjacent grids is l d , calculate the angle spectrum function and distance spectrum function of each grid point. According to the measured signal-to-noise ratio between the drone and each base station, the spectrum function is weighted to construct a new spectrum function, and then the function value of each point on the grid is calculated. Since the distance estimation is more accurate than the angle estimation, giving it a higher weight k>1 can increase the positioning accuracy, that is,

[0029]

[0030] S5. Because the signal space and the noise space are orthogonal, the larger the spectrum function value is, the closer it is to the true parameter. The grid point with the largest W(i) value is the final estimated result of the drone positioning. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 It is a schematic diagram of the system model of the present invention;

[0032] Figure 2 This is a schematic diagram of a multi-base station weighted fusion positioning coordinate grid of the present invention;

[0033] Figure 3 This is a comparison diagram of the positioning root mean square error between the present invention and the traditional single base station positioning method and the multi-base station average positioning method; DETAILED DESCRIPTION

[0034] The following is combined with the accompanying drawings and simulation examples to demonstrate the effectiveness and practicality of the present invention:

[0035] The simulation parameters are set as follows: the OFDM carrier frequency is set to 24 GHz, the frequency interval is 240 KHz, the number of carriers is 128, the number of symbols is 256, the planar antenna array is 4×4, the number of telepathic integrated base stations is 4, the maximum signal-to-noise ratio range of the communication channel is set to (-15, 0) dB, and the drone is located above all base stations. The number of points in the central grid is set to 5×5×5, the coordinate spacing between adjacent grid points is 0.3m, and the weight of the distance spectrum function is k=5. Repeating the simulation test can calculate the root mean square error between the estimated coordinates and the actual coordinates.

[0036] Depend on Figure 3 It can be seen that the root mean square error of the positioning of the present invention can be improved by 90% compared with the root mean square error of the traditional single base station positioning method, and can be improved by 50% compared with the multi-base station average positioning method. This proves that the present invention can provide significantly better accuracy than the traditional positioning algorithm by using the method of weighted fusion of multiple base stations according to the signal-to-noise ratio, and is a feasible and effective solution.

Claims

1. A synaesthesia integrated UAV positioning method based on multi-base station weighted fusion, characterized in that: The following steps are involved: S1. Construct a synaesthesia integrated positioning system, including N communication perception integrated base stations set on the ground and 1 UAV flying above each base station, each base station is equipped with a planar antenna array facing upward horizontally; initialize the real coordinates of the UAV and the synaesthesia integrated base station, each base station transmits an OFDM-based communication perception integrated waveform to the UAV, and receives the reflected wave from the UAV; S2. Each base station processes its transmit and receive signals respectively, and detects the incident angle and relative distance by comparing the changes in the transmit and receive signals to obtain the relative position of the drone; the receive signal matrix is ​​divided by the transmit signal matrix, and the expression of the response of the OFDM waveform corresponding to the μth symbol on the mth subcarrier of the nth carrier is as follows: Where R n With v n is the distance and radial velocity of the UAV in the direction corresponding to the nth base station, U n is the amplitude attenuation, η is the noise, the noise follows a Gaussian distribution with a mean of 0 and a variance of σ 2 ; Specifically: First, the elevation and azimuth angles are estimated. Each base station is equipped with a planar antenna array of size P × Q, which is placed parallel to the XOY plane and defines θ n and Respectively represent the azimuth and elevation of the received signal. If the drone is above the base station, the pitch angle The value range is (0,π / 2), the azimuth angle θ n The value range is (-π,π); the first unit on the upper left of the antenna array is taken as the reference unit, and the phase difference caused by the azimuth and elevation angles on different elements of the antenna array is detected. The phase difference between the antenna unit in the pth row and the qth column and the reference unit can be expressed as the steering vector is in The signals received by different base stations are expressed as: AND r,n =a n s r,n +N n where the vector s r,n is the nth base station echo signal matrix C n The rth row / column, N n represents additive white Gaussian noise; Receive angle signal matrix Y a,r,n The size is PQ×N s , and its autocorrelation matrix is ​​calculated as R a Perform eigenvalue decomposition and construct the noise subspace U a,r,n =U x (:,2:P×Q), where U x For R a The characteristic matrix is ​​arranged in descending order according to the size of the eigenvalues. The estimated values ​​of the elevation angle and the azimuth angle are obtained by searching for the maximum value of the spectrum function. Assuming that the communication channel is stable within the sampling time and the frequency response within the bandwidth is approximately the same, the estimation accuracy is improved by superimposing the spectrum functions of multiple subcarriers. On this basis, the following formula is obtained: Then find the peak value to reduce the influence of random noise, and record the high-precision estimation results of azimuth and elevation as and To estimate the distance, the process is similar to that of angle estimation. First, the characteristic matrix U of the distance autocorrelation matrix is ​​calculated. d,c,n , and then use the distance spectrum function superposition of multiple subcarriers to obtain a high-precision estimate, that is, By searching for peak values, high-precision estimates of distance and velocity are obtained. according to and The relative coordinates can be obtained. By knowing the position of the corresponding base station, each single base station can calculate a set of absolute coordinates of the drone, which can be recorded as S3, according to the measured signal-to-noise ratio between the UAV and each base station, a new absolute coordinate is synthesized, that is, S4, with new coordinates As the center, construct an n×n×n three-dimensional grid, and the coordinate spacing between adjacent grids is l d , calculate the angle spectrum function and distance spectrum function of each grid point; According to the measured signal-to-noise ratio between the UAV and each base station, the spectral function is weighted to construct a new spectral function, and then the function value of each point on the grid is calculated. Since the distance estimation is more accurate than the angle estimation, a higher weight k>1 is given to increase the positioning accuracy, that is, S5. Get the grid point with the largest W(i) value, which is the final estimated result of the drone positioning.

Images