Power line target detection method based on MIMO radar

Abstract

The invention discloses a power line target detection method based on an MIMO radar, and belongs to the technical field of radars, and the method comprises the steps: obtaining continuous multi-frame distance snapshot data through the MIMO radar, and carrying out the pulse compression of the obtained echo data; performing incoherent energy accumulation on the multi-frame distance snapshot data; a constant false alarm rate (CFAR) detection method is adopted to extract a distance gate of the suspected target; applying a high-resolution direction of arrival (DOA) estimation algorithm to the snapshot data in the range gate to obtain azimuth angle information; constructing a distance-azimuth two-dimensional image; and performing linear fitting on the target of the power line strong scattering point to obtain the distance between the power line and the radar and the orientation arrangement of the power line. According to the invention, the problems of insufficient detection performance and high false detection rate when facing a long-distance weak-reflection power line target in the prior art are solved. According to the invention, target signals can be enhanced, background noise can be suppressed, target detection and positioning precision can be improved, and the method has high sensitivity, high robustness and engineering feasibility.

Description

Technical Field

[0001] The present invention relates to the field of radar technology, and in particular to a power line target detection method based on MIMO radar. Background Art

[0002] Powerline inspection and obstacle avoidance are crucial tasks for power system maintenance and drone flight safety. With the continuous development of power systems and the increasing complexity of power lines, traditional inspection methods are no longer able to meet the requirements for efficient and safe operation and maintenance. Furthermore, the use of drones for power line inspections is becoming increasingly widespread, but their flight safety requires reliable obstacle avoidance technology. Therefore, developing a technology that can accurately detect the position and orientation of power lines in complex environments is crucial for the stable operation of power systems and the safe flight of drones.

[0003] Traditional power line detection methods rely primarily on visual sensors or lidar, which offer potential advantages in complex environments. However, their performance is limited in complex environments such as nighttime, dense fog, and rain and snow. While millimeter-wave radar offers advantages such as strong anti-interference capabilities, high resolution, and all-weather operation, accurately identifying the location and direction of power lines remains a technical challenge in power line monitoring.

[0004] Due to the relatively small radar cross-section of power lines, their reflection in millimeter-wave radar signals is weak, making them difficult to accurately identify. The environment surrounding power lines generates strong background noise and interference, affecting the accuracy of power line detection. Accurately identifying the location and direction of power lines in complex electromagnetic environments remains a technical challenge. Detection methods that rely on a single sensor fail to fully utilize the advantages of multiple sensors, resulting in insufficient detection accuracy and reliability. Summary of the Invention

[0005] The purpose of the present invention is to provide a power line target detection method based on MIMO radar, which can enhance target signals, suppress background noise, improve target detection and positioning accuracy, and has high sensitivity, high robustness and engineering feasibility, solving the problems raised in the above background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A power line target detection method based on MIMO radar, comprising:

[0008] Utilize MIMO radar to acquire continuous multi-frame range snapshot data and perform pulse compression processing on the acquired echo data;

[0009] Perform incoherent energy accumulation on multiple frames of range snapshot data to enhance the signal-to-noise ratio of the target echo signal;

[0010] On the distance-snapshot graph after energy accumulation, the constant false alarm rate (CFAR) detection method is used to extract the range gate of the suspected target.

[0011] Apply a high-resolution direction of arrival (DOA) estimation algorithm to the snapshot data within the range gate to obtain the azimuth information of all target points;

[0012] Construct a range-azimuth two-dimensional image based on all target range and azimuth information;

[0013] Perform straight line fitting on all targets that may be strong scattering points of power lines to obtain the distance between the power lines and the radar and the azimuth arrangement of the power lines.

[0014] Preferably, the MIMO radar is composed of multiple transmitting array elements and multiple receiving array elements, and is used to obtain high-resolution azimuth estimation capability by synthesizing a virtual array.

[0015] Preferably, using a MIMO radar to acquire continuous multiple frames of range snapshot data and performing pulse compression processing on the acquired echo data includes:

[0016] MIMO radar transmits frequency modulated continuous wave signals, and multiple receiving channels receive echo signals reflected from the target and perform pulse compression processing in the frequency domain through matched filtering;

[0017] Among them, the matched filter of the echo signal adopts a frequency domain amplitude weighted window and selects a Taylor weighted window with a maximum sidelobe level of -80dB to reduce the impact of the sidelobe on adjacent distance units while maintaining a high signal-to-noise ratio performance.

[0018] Preferably, the non-coherent energy accumulation is performed on the multiple frames of range snapshot data by point-by-point accumulation of the squared modulus values of the echo signals of multiple consecutive frames;

[0019] For each range unit in each frame of the range-snapshot image, the square of the modulus length of its complex echo signal is taken, that is:

[0020] P n (r)=|s n (r)∣ 2

[0021] Among them, s n (r) represents the complex echo signal of the r-th range unit in the n-th frame, P n (r) represents the power value at that point;

[0022] Add the power values of the corresponding distance unit positions in N frames point by point to form the energy accumulation image P acc (r), where the energy accumulation image P acc(r) is obtained by the following formula:

[0023]

[0024] Among them, P acc (r) represents the energy accumulation image.

[0025] Preferably, the constant false alarm rate (CFAR) detection method adopts a sliding window mechanism to dynamically set the detection threshold for adaptively responding to the noise levels of different distance units. The constant false alarm rate (CFAR) detection method is used to count the maximum energy in the reference unit, screen out suspected target units whose detection units are greater than the maximum value, and record their corresponding distance positions to form a set of target range gates.

[0026] Preferably, the high-resolution direction of arrival (DOA) estimation algorithm is a MUSIC algorithm.

[0027] Preferably, a high-resolution direction of arrival (DOA) estimation algorithm is applied to the snapshot data within the range gate to obtain the azimuth information of all target points, including:

[0028] In each range gate marked as a suspected target by the constant false alarm rate (CFAR) detection, the MIMO channel echo data corresponding to the range unit in all snapshot frames is extracted to form a snapshot data matrix X(r)∈C M×L , where M is the number of virtual array channels and L is the number of snapshot frames;

[0029] Among them, the forward and backward spatial smoothing technology is used, and the receiving array element is a uniform linear array. If there are M array elements in total, the number of targets D is generally smaller than the number of receiving array elements. The received snapshot data matrix is:

[0030]

[0031] Among them, L is the number of snapshot frames, is the received data of the first frame;

[0032] The entire array is divided into K=M-P+1 sub-arrays, each sub-array contains P array elements, then the k-th forward sub-array position is:

[0033]

[0034] Similarly, the corresponding backward submatrix is:

[0035]

[0036] in, is the anti-diagonal identity matrix, * represents the complex conjugate;

[0037] All forward and backward sub-matrices are averaged to construct a smoothed covariance matrix:

[0038]

[0039] make The formula can be simplified as:

[0040]

[0041] Preferably, the covariance matrix after forward and backward smoothing is restructured using Toeplitz, including:

[0042] Assume t d represents the diagonal mean value with difference d, and the reconstruction matrix is defined as follows:

[0043]

[0044] in, This is the final constructed Toeplitz covariance matrix, which has the structural characteristics of constant diagonal lines.

[0045] Perform eigendecomposition on the covariance matrix and divide the signal subspace and noise subspace into:

[0046]

[0047] Construct a spectral function using the noise subspace:

[0048]

[0049] The above algorithm can obtain more accurate and stable azimuth estimation results.

[0050] Preferably, a range-azimuth two-dimensional image is constructed based on all target range and azimuth information, including:

[0051] The distance information r of each scattering target point is jointly mapped with its corresponding azimuth information θ to construct a range-azimuth two-dimensional image;

[0052] The target distance information r and azimuth information θ are converted into spatial position coordinates in a rectangular coordinate system, and the conversion relationship is:

[0053] x=r·cos(θ),y=r·sin(θ)

[0054] By performing the above coordinate transformation on all target points, the mapping of the distance-azimuth diagram to a rectangular two-dimensional space diagram is completed, which is used to provide an intuitive and structured data basis for the subsequent geometric fitting of the power line trajectory.

[0055] Preferably, performing linear fitting on all targets that may be strong scattering points of power lines to obtain the distance between the power lines and the radar and the azimuth arrangement of the power lines includes:

[0056] Target point screening: extract points with significant scattering intensity from the range-azimuth two-dimensional image as a set of candidate target points;

[0057] RANSAC fitting: Use the RANSAC algorithm to fit a straight line model to the target point set. During the iteration process, the vertical distance from the point to the line is minimized and outliers are excluded to achieve robust linear structure recognition.

[0058] Power line parameter extraction: The fitted straight line parameters are used to determine the direction angle and distance of the power line and estimate its tilt in the image space, providing geometric reference information for subsequent navigation, obstacle avoidance, or monitoring.

[0059] Compared with the prior art, the present invention has the following beneficial effects:

[0060] The present invention integrates multi-frame incoherent energy accumulation and high-resolution directional angle of arrival estimation technology, enhances the target echo energy through multi-frame radar data, improves the overall signal-to-noise ratio, adopts the constant false alarm rate CFAR detection method to extract the distance unit of the suspected target, and introduces the MUSIC algorithm for high-resolution azimuth estimation to achieve range-azimuth two-dimensional imaging and power line positioning. It can enhance the target signal, suppress background noise, and improve the target detection and positioning accuracy. It makes full use of the consistency characteristics of the target in time and space dimensions, significantly improving the detection probability and directional accuracy of low RCS targets. It can be widely used in scenarios such as power tower status perception and drone obstacle avoidance navigation, and has high sensitivity, high robustness and engineering feasibility. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 Flowchart of the power line target detection method based on MIMO radar of the present invention;

[0062] Figure 2 Schematic diagram of the processing result of the constant false alarm rate (CFAR) detection range gate of the present invention;

[0063] Figure 3 This is a schematic diagram of the results of azimuth estimation using the MUSIC algorithm of the present invention;

[0064] Figure 4 Schematic diagram of the construction result of the range-azimuth two-dimensional image of the present invention;

[0065] Figure 5 Schematic diagram of the result of straight line fitting of the present invention. DETAILED DESCRIPTION

[0066] 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, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0067] To address the existing radar detection methods that typically use single-frame CFAR detection and FFT spectrum analysis, which are limited by energy loss and insufficient directional resolution in low signal-to-noise ratio environments, it is difficult to effectively extract the echo characteristics of distant power line targets. In particular, when the power line RCS is small or the attitude changes drastically, the detection rate drops significantly. When facing distant, weakly reflecting power line targets, the detection performance is insufficient and the false detection rate is high. Please refer to Figure 1-Figure 5 , this embodiment provides the following technical solutions:

[0068] A power line target detection method based on MIMO radar, comprising:

[0069] MIMO radar is used to obtain continuous multi-frame distance snapshot data, and the obtained echo data is pulse compressed.

[0070] In this embodiment, the MIMO radar is composed of multiple transmitting array elements and multiple receiving array elements, and is used to obtain high-resolution azimuth estimation capability by synthesizing a virtual array.

[0071] It should be noted that the radar system first transmits a frequency-modulated continuous wave signal, and multiple receiving channels receive the echo signal reflected from the target, and perform pulse compression processing in the frequency domain through matched filtering. Its performance is largely limited by the level of range sidelobes. Since the difference in range products between scatterers in different range units is significantly greater than the difference in scatterers in the same range unit, the existence of range sidelobes will significantly interfere with the detection of weak targets, especially against backgrounds such as the ground, which can easily drown out the echo signals of power line target points.

[0072] Therefore, in this embodiment, a frequency domain amplitude weighted window is used for the matched filter of the echo signal to suppress the influence of the side lobes and reduce the signal-to-noise ratio loss of the main lobe. In view of the weak echo characteristics of the power line target and the system's requirement for low side lobes, a Taylor weighted window with a maximum side lobe level of -80dB is selected to effectively reduce the influence of the side lobes on adjacent distance units, while maintaining a high signal-to-noise ratio performance as much as possible to enhance the overall target detection capability of the system.

[0073] Specifically, by making full use of the array structure and imaging capabilities of MIMO radar, combined with the position stability and scattering characteristics of power lines in space, a joint detection algorithm of "multi-frame incoherent energy accumulation + high-resolution directional angle of arrival (DOA) estimation" is designed to enhance target signals, suppress background noise, and improve target detection and positioning accuracy. The power line detection technology based on MIMO radar high-angle resolution technology has a certain suppression ability against interference due to the orthogonality of the transmitted waveform. The virtual antenna array expansion is used to significantly improve the angular resolution. Combined with multi-frame accumulation and high-resolution direction estimation, the detectability of weak target signals is enhanced, effectively improving the detection rate of low RCS power line targets in complex environments.

[0074] Incoherent energy accumulation is performed on multiple frames of range snapshot data to enhance the signal-to-noise ratio of the target echo signal.

[0075] It should be noted that in order to further improve the detectability of echo signals from weak targets (such as power lines), incoherent energy accumulation processing is performed on the echo data collected by the radar system in multiple consecutive snapshot frames.

[0076] In this embodiment, for each range unit in each frame of the range-snapshot image, the square of the modulus length of the complex echo signal is calculated, that is:

[0077] P n (r)=|s n (r)∣ 2

[0078] Among them, s n (r) represents the complex echo signal of the r-th range unit in the n-th frame, P n (r) represents the power value at that point;

[0079] Add the power values of the corresponding distance unit positions in N frames point by point to form the energy accumulation image P acc (r), where the energy accumulation image P acc (r) is obtained by the following formula:

[0080]

[0081] Among them, P acc (r) represents the energy accumulation image;

[0082] Therefore, incoherent energy accumulation can significantly improve the signal-to-noise ratio, and the improvement is theoretically proportional to the number of frames N without introducing phase interference. Therefore, it is particularly suitable for scenarios where the target echo amplitude is stable but the phase is random, such as power line scattering echo.

[0083] On the distance-snapshot graph after energy accumulation, the constant false alarm rate (CFAR) detection method is used to extract the range gate of the suspected target.

[0084] It should be noted that after completing incoherent energy accumulation, the resulting range-snapshot image reflects the energy distribution of each range cell in the radar field of view. To automatically extract the location of potential targets from this image, a constant false alarm rate (CFAR) detection algorithm is used to process the image. This algorithm calculates the maximum energy value in a reference cell, selects suspected target cells with detection cells greater than this maximum value, and records their corresponding range positions to form a set of target range gates. Therefore, CFAR detection enables reliable extraction of power line echoes in complex backgrounds, effectively suppressing false alarms and providing an accurate range index for subsequent high-resolution direction of arrival angle estimation.

[0085] In this embodiment, the constant false alarm rate (CFAR) detection method adopts a sliding window mechanism and dynamically sets a detection threshold to adaptively cope with noise levels of different distance units.

[0086] A high-resolution direction of arrival (DOA) estimation algorithm is applied to the snapshot data within the range gate to obtain the azimuth information of all target points.

[0087] It should be noted that after completing the constant false alarm rate (CFAR) detection and obtaining the range gates of multiple suspected targets, the directional information of these targets is further extracted. Based on the angular resolution capability provided by the MIMO radar system, a high-resolution directional angle of arrival estimation algorithm is used to accurately measure the azimuth of the target point.

[0088] In this embodiment, the high-resolution direction of arrival (DOA) estimation algorithm is a MUSIC algorithm.

[0089] In this embodiment, within each range gate marked as a suspected target by the constant false alarm rate (CFAR) detection, the MIMO channel echo data corresponding to the range unit in all snapshot frames is extracted to form a snapshot data matrix X(r)∈C M×L , where M is the number of virtual array channels and L is the number of snapshot frames.

[0090] In order to overcome the correlation between incident signals in a multi-target signal environment and solve the problem that the traditional covariance matrix construction method will lead to a decrease in matrix rank, the forward and backward spatial smoothing technology is used, which has stronger resolution and better anti-coherence performance than ordinary smoothing technology.

[0091] The receiving array element is a uniform linear array. If there are M array elements in total, the number of targets D is generally smaller than the number of receiving array elements. The received snapshot data matrix is:

[0092]

[0093] Among them, L is the number of snapshot frames, is the received data of the first frame;

[0094] The entire array is divided into K=M-P+1 sub-arrays, each sub-array contains P array elements, then the k-th forward sub-array position is:

[0095]

[0096] Similarly, the corresponding backward submatrix is:

[0097]

[0098] in, is the anti-diagonal identity matrix, * represents the complex conjugate;

[0099] All forward and backward sub-matrices are averaged to construct a smoothed covariance matrix:

[0100]

[0101] make The formula can be simplified as:

[0102]

[0103] To further enhance the structural consistency of the array covariance matrix and improve the resolution and robustness of high-resolution DOA estimation, the covariance matrix after forward and backward smoothing is restructured using the Toeplitz method, including:

[0104] Assume t d represents the diagonal mean value with difference d, and the reconstruction matrix is defined as follows:

[0105]

[0106] in, This is the final constructed Toeplitz covariance matrix, which has the structural characteristics of constant diagonal lines.

[0107] Perform eigendecomposition on the covariance matrix and divide the signal subspace and noise subspace into:

[0108]

[0109] Construct a spectral function using the noise subspace:

[0110]

[0111] The above algorithm can obtain more accurate and stable azimuth estimation results, especially when the target signal strength fluctuates or is highly coherent with each other, while still maintaining high resolution and robustness.

[0112] Among them, the angle estimation of the range gate where the target is located will produce several peaks. Multiple peaks can be selected and retained as pseudo target points to prevent missing the strong Bragg scattering points of the power lines.

[0113] A range-azimuth two-dimensional image is constructed based on all target range and azimuth information.

[0114] In this embodiment, a range-azimuth two-dimensional image is constructed based on all target range and azimuth information, including:

[0115] The distance information r of each scattering target point is jointly mapped with its corresponding azimuth information θ to construct a range-azimuth two-dimensional image;

[0116] The target distance information r and azimuth information θ are converted into spatial position coordinates in a rectangular coordinate system, and the conversion relationship is:

[0117] x=r·cos(θ),y=r·sin(θ)

[0118] By performing the above coordinate transformation on all target points, the mapping of the distance-azimuth diagram to a rectangular two-dimensional space diagram is completed, which is used to provide an intuitive and structured data basis for the subsequent geometric fitting of the power line trajectory.

[0119] Perform straight line fitting on all targets that may be strong scattering points of power lines to obtain the distance between the power lines and the radar and the azimuth arrangement of the power lines.

[0120] It should be noted that in the constructed range-azimuth two-dimensional image, the power line target usually appears as a group of strong scattering points arranged continuously along a specific direction. In order to identify the spatial distribution characteristics of the power lines, a robust straight line fitting algorithm is used for processing. This can effectively eliminate outliers and confirm the straight direction of the power lines.

[0121] In this embodiment, a linear fitting is performed on all targets that may be strong scattering points of the power lines to obtain the distance between the power lines and the radar and the azimuth arrangement of the power lines, including:

[0122] Target point screening: extract points with significant scattering intensity from the range-azimuth two-dimensional image as a set of candidate target points;

[0123] RANSAC fitting: Use the RANSAC algorithm to fit a straight line model to the target point set. During the iteration process, the vertical distance from the point to the line is minimized and outliers are excluded to achieve robust linear structure recognition.

[0124] Power line parameter extraction: The fitted straight line parameters are used to determine the direction angle and distance of the power line and estimate its tilt in the image space, providing geometric reference information for subsequent navigation, obstacle avoidance, or monitoring.

[0125] Therefore, it can be deployed on fixed radar monitoring equipment (such as on power poles) to realize real-time monitoring of power line swing, tilt and other conditions; it can also be integrated into mobile platforms (such as drones, power inspection robots, etc.) to identify the location of power lines in complex environments and realize obstacle avoidance and navigation functions.

[0126] In summary, by integrating multi-frame incoherent energy accumulation and high-resolution directional angle of arrival estimation technology, the target echo energy is enhanced through multi-frame radar data, the overall signal-to-noise ratio is improved, the constant false alarm rate CFAR detection method is used to extract the distance unit of the suspected target, and the MUSIC algorithm is introduced for high-resolution azimuth estimation to achieve range-azimuth two-dimensional imaging and power line positioning. It can enhance the target signal, suppress background noise, and improve the target detection and positioning accuracy. It makes full use of the consistency characteristics of the target in time and space dimensions, significantly improving the detection probability and directional accuracy of low RCS targets. It can be widely used in scenarios such as power tower status perception and drone obstacle avoidance navigation, and has high sensitivity, high robustness and engineering feasibility.

[0127] 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 "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0128] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A power line target detection method based on MIMO radar, characterized in that: include: Utilize MIMO radar to acquire continuous multi-frame range snapshot data and perform pulse compression processing on the acquired echo data; Perform incoherent energy accumulation on multiple frames of range snapshot data to enhance the signal-to-noise ratio of the target echo signal; On the distance-snapshot graph after energy accumulation, the constant false alarm rate (CFAR) detection method is used to extract the range gate of the suspected target. Apply a high-resolution direction of arrival (DOA) estimation algorithm to the snapshot data within the range gate to obtain the azimuth information of all target points; Construct a range-azimuth two-dimensional image based on all target range and azimuth information; Perform straight line fitting on all targets that may be strong scattering points of power lines to obtain the distance between the power lines and the radar and the azimuth arrangement of the power lines.

2. The power line target detection method based on MIMO radar according to claim 1, characterized in that: The MIMO radar is composed of multiple transmitting array elements and multiple receiving array elements, and is used to obtain high-resolution azimuth estimation capability by synthesizing a virtual array.

3. The power line target detection method based on MIMO radar according to claim 1, characterized in that: Utilize MIMO radar to acquire continuous multi-frame range snapshot data and perform pulse compression processing on the acquired echo data, including: MIMO radar transmits frequency modulated continuous wave signals, and multiple receiving channels receive echo signals reflected from the target and perform pulse compression processing in the frequency domain through matched filtering; Among them, the matched filter of the echo signal adopts a frequency domain amplitude weighted window and selects a Taylor weighted window with a maximum sidelobe level of -80dB to reduce the impact of the sidelobe on adjacent distance units while maintaining a high signal-to-noise ratio performance.

4. The method for detecting power line targets based on MIMO radar according to claim 1, wherein: The method of performing non-coherent energy accumulation on multiple frames of range snapshot data is point-by-point accumulation of the squared modulus values of the echo signals of multiple consecutive frames; For each range unit in each frame of the range-snapshot image, the square of the modulus length of its complex echo signal is taken, that is: P n (r)=∣s n (r)∣ 2 Among them, s n (r) represents the complex echo signal of the r-th range unit in the n-th frame, P n (r) represents the power value at that point; Add the power values of the corresponding distance unit positions in N frames point by point to form the energy accumulation image P acc (r), where the energy accumulation image P acc (r) is obtained by the following formula: Among them, P acc (r) represents the energy accumulation image.

5. The method for detecting power line targets based on MIMO radar according to claim 1, wherein: The CFAR detection method uses a sliding window mechanism to dynamically set the detection threshold to adaptively respond to the noise levels of different distance cells. The CFAR detection method uses the maximum energy value in the reference cell to filter out suspected target cells with detection cells greater than the maximum value, and records their corresponding distance positions to form a set of target range gates.

6. The method for detecting power line targets based on MIMO radar according to claim 1, wherein: The high-resolution direction of arrival (DOA) estimation algorithm is the MUSIC algorithm.

7. The method for detecting power line targets based on MIMO radar according to claim 1, wherein: Apply a high-resolution Direction of Arrival (DOA) estimation algorithm to the snapshot data within the range gate to obtain the azimuth information of all target points, including: In each range gate marked as a suspected target by the constant false alarm rate (CFAR) detection, the MIMO channel echo data corresponding to the range unit in all snapshot frames is extracted to form a snapshot data matrix X(r)∈C M×L , where M is the number of virtual array channels and L is the number of snapshot frames; Among them, the forward and backward spatial smoothing technology is used, and the receiving array element is a uniform linear array. If there are M array elements in total, the number of targets D is generally smaller than the number of receiving array elements. The received snapshot data matrix is: Among them, L is the number of snapshot frames, is the received data of the first frame; The entire array is divided into K=M-P+1 sub-arrays, each sub-array contains P array elements, then the k-th forward sub-array position is: Similarly, the corresponding backward submatrix is: in, is the anti-diagonal identity matrix, * represents the complex conjugate; All forward and backward sub-matrices are averaged to construct a smoothed covariance matrix: make The formula can be simplified as:

8. The method for detecting power line targets based on MIMO radar according to claim 7, wherein: Perform Toeplitz restructuring on the covariance matrix after forward and backward smoothing, including: Assume t d represents the diagonal mean value with difference d, and the reconstruction matrix is defined as follows: in, This is the final constructed Toeplitz covariance matrix, which has the structural characteristics of constant diagonal lines. Perform eigendecomposition on the covariance matrix and divide the signal subspace and noise subspace into: Construct a spectral function using the noise subspace: Get the azimuth estimation result.

9. The method for detecting power line targets based on MIMO radar according to claim 1, wherein: Construct a range-azimuth two-dimensional image based on all target range and azimuth information, including: The distance information r of each scattering target point is jointly mapped with its corresponding azimuth information θ to construct a range-azimuth two-dimensional image; The target distance information r and azimuth information θ are converted into spatial position coordinates in a rectangular coordinate system, and the conversion relationship is: x=r·cos(θ),y=r·sin(θ) By performing the above coordinate transformation on all target points, the mapping of the distance-azimuth diagram to a rectangular two-dimensional space diagram is completed, which is used to provide an intuitive and structured data basis for the subsequent geometric fitting of the power line trajectory.

10. The method for detecting power line targets based on MIMO radar according to claim 1, wherein: Perform linear fitting on all targets that may be strong scattering points of power lines to obtain the distance between the power lines and the radar and the azimuth distribution of the power lines, including: Target point screening: extract points with significant scattering intensity from the range-azimuth two-dimensional image as a set of candidate target points; RANSAC fitting: Use the RANSAC algorithm to fit a straight line model to the target point set. During the iteration process, the vertical distance from the point to the straight line is minimized and outliers are excluded. Power line parameter extraction: The fitted straight line parameters are used to determine the direction angle and distance of the power line and estimate its tilt in the image space, providing geometric reference information for subsequent navigation, obstacle avoidance, or monitoring.

Images