High-precision multi-station target detection method fusing time delay and frequency shift

Abstract

This invention discloses a high-precision multi-station target detection method that integrates time delay and frequency shift, relating to the field of signal processing technology. The implementation process is as follows: (1) Divide the space into subspaces uniformly according to the characteristics of the measurement space; (2) Calculate the theoretical time delay and theoretical Doppler frequency shift of targets at different speeds at the center position of each subspace relative to the radar, and save them as a dictionary matrix; (3) Utilize the time delay and Doppler frequency shift of moving targets in space relative to each radar, and obtain the possible weights of the targets in each subspace through a multi-task Bayesian compressed sensing algorithm; (4) Determine the location of the target based on the maximum weight, and detect and locate the target. Simulation data and experimental results demonstrate the effectiveness and superiority of the target detection method of this invention. The method of this invention uses time delay and Doppler frequency shift features to jointly estimate the target position. Simulation results show that it is better than the commonly used time delay-only estimation method, and can be used for space exploration of UAVs, aircraft, etc.

Description

Technical Field

[0001] This invention relates to the field of signal processing technology, specifically to a high-precision multi-station target detection method that integrates delay and frequency shift. Background Technology

[0002] In radar signal processing, target detection and localization methods have always been a key research focus. Conventional radar signal processing uses only time delay for target localization, without frequency shift. However, with continuous technological advancements, the acquisition and utilization of multi-dimensional information has become increasingly important. Since the 1930s, the Doppler effect has been widely applied in various fields, including the electromagnetic field. Radar transmits signals to a target while simultaneously receiving the reflected signals. The true distance between the radar and the target can be estimated based on the time difference between the transmitted and received signals. In practical applications, since most targets to be detected by radar are in constant motion, the frequency received by the radar will shift from the transmitted signal. This frequency shift is caused by the Doppler effect. As the frequency of Doppler radar increases, its accuracy also improves. The rapid development of signal processing technology has significantly improved speed measurement performance, leading to the widespread application of various Doppler speed measuring instruments developed based on this phenomenon. This has also given rise to the fastest tracking technology and lane-keeping technology. In practical applications, many technological means utilize the Doppler effect, with target detection being a crucial application.

[0003] Compressed sensing theory, as an emerging data sampling and processing method, is unique in that it directly obtains the effective information in the signal through undersampling, reconstructs the sampled signal, and finally recovers the original signal. This method does not require a large amount of data, avoids excessive data sampling, improves storage efficiency, and reduces the pressure on hardware devices.

[0004] Therefore, this paper proposes a high-precision multi-station target detection method that integrates time delay and frequency shift to address the above problems. Finding an accurate and efficient target detection method has always been a research focus. By utilizing the Doppler effect of multiple stations and using time delay and Doppler frequency shift information, the efficiency and accuracy of target localization can be effectively improved. Summary of the Invention

[0005] The purpose of this invention is to provide a high-precision multi-station target detection method that integrates time delay and frequency shift, thereby addressing the problem in the background art that existing target detection methods only use time delay for target localization and do not utilize frequency shift, leaving significant room for improvement in the acquisition and utilization of multi-dimensional information. To achieve the above objective, this invention provides the following technical solution: a high-precision multi-station target detection method that integrates time delay and frequency shift, comprising the following steps:

[0006] A three-dimensional coordinate system is established in the detection space, which includes a moving target to be located, and the position of the emission source is located at... The location of the nth receiver is located at , Both the transmitter and receiver are stationary, and their locations are known.

[0007] Step 1: Divide the space into subspaces evenly based on the characteristics of the measured space.

[0008]

[0009] This invention assumes that the size of the scene to be detected is The scene to be detected is divided into blocks, each block being the size of a unit. Here, L, W, and H are divisible by l, w, and h respectively. The entire space is divided into M subspaces, and a moving target may have K velocities. Therefore, one radar corresponds to... Subspace;

[0010] Step 2: Calculate the theoretical delay and theoretical Doppler shift of targets at different speeds at the center position of each subspace from each radar, and save them as a dictionary matrix;

[0011] (a) The actual compressed sensing process is for each The observation vectors of each subspace are obtained by observing the space of size individually. The state vector of the moving target in any subspace is: It includes the position in a three-dimensional Cartesian coordinate system. and speed ;

[0012] (b) Calculate and save the theoretical delay between the moving target's position at the center of each subspace and each radar, using the following formula:

[0013]

[0014] Where C is the speed of light;

[0015] (c) The theoretical Doppler frequency shift for moving targets at different velocities at the center position of each subspace is calculated using the following formula:

[0016]

[0017] in, Where λ is the wavelength of the transmitted signal, and C is the speed of light;

[0018] (d) Integrating theoretical delay and Doppler frequency shift into The dictionary matrix of each radar is obtained:

[0019]

[0020] Step 3: Using the time delay and Doppler frequency shift of the moving target in space relative to the radar, the weights of the target's possible existence in each subspace are obtained through a multi-task Bayesian compressed sensing algorithm.

[0021] (a) The calculated time delay and Doppler frequency shift of the moving target in space relative to the radar are measured values, denoted as:

[0022]

[0023] (b) Unknown sparse vectors in the target state space It can be obtained by solving the sparse solution of the linear equation of the following expression.

[0024]

[0025] in It is noise generated during the observation process, with a variance of For a zero-mean Gaussian random variable, the parameters can be obtained from the above formula. and The likelihood function is:

[0026]

[0027] Assumption Each element in the set follows a Gaussian prior distribution with a mean of zero:

[0028]

[0029] in It is a Gaussian density function with zero mean. yes The j-th element, and Both can be represented by a gamma distribution:

[0030]

[0031]

[0032] Setting a=b=0, c=d=0, we obtain the result using Bayes' criterion. The posterior probability density function is:

[0033]

[0034] Where the mean and variance are...

[0035]

[0036]

[0037]

[0038]

[0039]

[0040] in, yes The j-th element, yes Obtain the j-th diagonal element and set the parameters appropriately. The above formula is recursively updated until the convergence criterion is met, and the mean is obtained. and variance The mean is used as the pair of vectors The estimated value can be obtained by effectively utilizing group sparsity through a multi-task Bayesian compressed sensing algorithm. To further obtain the target location The estimate;

[0041] Step 4: Determine the location of the target based on the maximum weight, and then detect and locate the target.

[0042] In obtaining the sparse reconstruction estimation vector … Then, the subspace representing the maximum value of the sum of the sparse reconstructed vectors is taken as the estimated state, and the position pointed to by this subspace is taken as the estimated position. The average distance difference between the estimated position and the actual position is calculated.

[0043] Compared with the prior art, the beneficial effects of the present invention are:

[0044] 1. This high-precision multi-station target detection method that integrates time delay and frequency shift constructs a dictionary matrix by fusing time delay and Doppler frequency shift. The target reconstruction can be directly completed through the time delay and Doppler frequency shift characteristics of the echo, and the positioning result can be obtained, which reduces the amount of computation and improves the measurement accuracy compared with positioning using only time delay.

[0045] 2. This high-precision multi-station target detection method that integrates delay and frequency shift applies compressed sensing theory, and the sampling rate can be much lower than the Nyquist sampling rate, reducing the storage space and computational load of the acquired signals;

[0046] 3. This high-precision multi-station target detection method that integrates delay and frequency shift can locate targets at any speed. It uses information from multiple stations and performs better when the signal-to-noise ratio is the same. The more receiving stations there are, the higher the efficiency. Attached Figure Description

[0047] Figure 1 This is a schematic diagram of the process of the present invention;

[0048] Figure 2 This is a schematic diagram of the simulation scenario of the present invention;

[0049] Figure 3 A comparison diagram showing the results of localization using delay and Doppler frequency shift features versus using delay features alone, based on embodiments of the present invention;

[0050] Figure 4 The figure shows a comparison of localization results using delay and Doppler frequency shift features versus using only delay features, based on embodiments of the present invention. Detailed Implementation

[0051] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. The described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0052] See attached document Figure 2 A three-dimensional coordinate system is established in the detection space, which includes a moving target to be located, and the position of the emission source is located at... The locations of the three receivers are respectively located at , , Given that both the transmitter and receiver are stationary and their positions are known, the task of this invention, a high-precision multi-station target detection method fusing delay and frequency shift, is to estimate the target's position using the delay and Doppler frequency shift of the target received from multiple stations. The method includes the following steps:

[0053] Step 1: Divide the space into subspaces evenly based on the characteristics of the measurement space. This invention assumes the size of the scene to be detected is... The scene to be detected is divided into blocks, each block being the size of a unit. The speed of the moving target may be 10m / s, 20m / s, or 30m / s;

[0054] Step 2: Calculate the theoretical time delay and theoretical Doppler frequency shift of targets at different speeds at the center position of each subspace from each radar, and save them as a dictionary matrix;

[0055] (a) is for each The observation vectors of each subspace are obtained by observing the space of size individually. The state vector of the moving target in any subspace is: It includes the position in a three-dimensional Cartesian coordinate system. and speed .

[0056] (b) Calculate and save the theoretical delay between the moving target's position at the center of each subspace and each radar, using the following formula:

[0057]

[0058] (c) The theoretical Doppler frequency shift of a moving target at different velocities at the center position of each subspace is calculated using the following formula:

[0059]

[0060] in, Where λ is the wavelength of the transmitted signal, and C is the speed of light;

[0061] (d) Integrating theoretical delay and Doppler frequency shift into The dictionary matrix of each radar is obtained:

[0062]

[0063]

[0064]

[0065] Step 3: Using the delay and Doppler frequency of the moving target in space relative to the radar, the weights of the target's possible existence in each subspace are obtained through a multi-task Bayesian compressed sensing algorithm.

[0066] (a) The calculated time delay and Doppler frequency of the moving target in space relative to the radar are measured values, denoted as:

[0067]

[0068] (b) Unknown sparse vectors in the target state space It can be obtained by solving the sparse solution of the linear equation of the following expression.

[0069]

[0070] in It is noise generated during the observation process, with a variance of A zero-mean Gaussian random variable;

[0071] We can obtain this through Bayes' criterion. The posterior probability density function is:

[0072]

[0073] Where the mean and variance are...

[0074]

[0075]

[0076]

[0077]

[0078]

[0079] in, yes The j-th element, yes Obtain the j-th diagonal element and set the parameters appropriately. The above formula is recursively updated until the convergence criterion is met, and the mean is obtained. and variance The mean is used as the pair of vectors The estimated value can be obtained by effectively utilizing group sparsity through a multi-task Bayesian compressed sensing algorithm. To further obtain the target location The estimate;

[0080] Step 4: Determine the location of the target based on the maximum weight, and then detect and locate the target;

[0081] In obtaining the sparse reconstruction estimation vector , , Then, the state subspace represented by the maximum value of the sum of the sparse reconstruction vectors is taken as the estimated state, and the position pointed to by this subspace is taken as the estimated position. The average distance difference between the estimated position and the actual position is calculated.

[0082] To verify the effectiveness and superiority of the proposed state estimation method that integrates time delay and Doppler features, simulation experiments were conducted, and performance comparisons were performed. The experiments simulated two scenarios to verify the advantages of the invention. Scenario 1: The target is located at the center of the subspace. In this case, the measurement matrix data and the dictionary matrix data are relatively consistent. Noise with different signal-to-noise ratios was added to the measurement matrix to simulate measurement errors. Figure 3 The estimation method using fused delay and Doppler frequency shift features obtained in this invention performs better than using delay estimation alone, and can achieve target estimation at lower signal-to-noise ratios. Case 2: The target is not at the center of the subspace. Setting the target's position to be 2m away from the center of the receiving station, we examine whether the target can be located to the nearest subspace center. In this case, there is a significant difference between the measurement matrix data and the dictionary matrix data. (The remaining text appears to be incomplete and requires further context.) Figure 4 The method of the present invention can achieve better results. The positioning accuracy of the method of the present invention is higher than that of positioning using only delay at various signal-to-noise ratios.

[0083] In summary, this invention discloses a high-precision multi-station target detection method that integrates delay and frequency shift. Addressing the sparsity of targets in space, the method analyzes the delay and Doppler information received by each station from the signals received by multiple stations. Sparse signal estimation is obtained through signal reconstruction using Bayesian compressed sensing. By reasonably dividing the subspace and setting the estimation velocity interval, the estimation performance can be optimized. It should be understood that the above description is only the most preferred embodiment of this invention and is not intended to limit the invention. Any method that does not exceed the protection scope described in the claims of this invention should be within the protection scope of this invention.

[0084] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

Claims

1. A high-precision multi-station target detection method integrating time delay and frequency shift, characterized in that: Includes the following steps: Step 1: Divide the space into subspaces evenly based on the characteristics of the measured space; Step 2: Calculate the theoretical delay and theoretical Doppler shift of the range radar when targets at different speeds are at the center position of each subspace, and save them as a dictionary matrix; Step 3: Utilize the time delay and Doppler frequency shift of moving targets in space relative to each radar to obtain the weights of the targets that may exist in each subspace through a multi-task Bayesian compressed sensing algorithm; Step 4: Determine the location of the target based on the maximum weight, and then detect and locate the target.

2. The high-precision multi-station target detection method that integrates delay and frequency shift according to claim 1, characterized in that: Based on the assumption in step one that the size of the scene to be measured is... The scene to be measured is divided into blocks, each block being the size of a unit. The entire space is divided into M subspaces, and a moving target may have K velocities, therefore one radar corresponds to... Subspace.

3. The high-precision multi-station target detection method that integrates delay and frequency shift according to claim 2, characterized in that: The calculation steps based on step two are as follows: The actual compressed sensing process is for each The observation vectors of each subspace are obtained by observing the space of size individually. The state vector of the moving target in any subspace is: Includes position in a three-dimensional Cartesian coordinate system and speed ; The formula used to calculate and save the theoretical delay between the moving target's position at the center of each subspace and each radar is as follows: ； Where C is the speed of light; The formula used to calculate the theoretical Doppler frequency shift of a moving target at different speeds at the center position of each subspace is: ； in, Where λ is the wavelength of the transmitted signal, and C is the speed of light. For carrier frequency; Integrating theoretical delay and Doppler frequency shift into Obtain the dictionary matrix for each radar: ； 4. The high-precision multi-station target detection method that integrates delay and frequency shift according to claim 3, characterized in that: The calculation steps based on step three are as follows: The calculated time delay and Doppler frequency of the moving target in space relative to the radar are the measured values, which constitute the measurement matrix: Unknown sparse vectors in the target state space We obtain the following by solving the sparse solution of the linear equation: ； in, It is noise generated during the observation process, with a variance of The zero-mean Gaussian random variable, the parameter is obtained through the above formula. and The likelihood function is: ； Assumption Each element in the set follows a Gaussian prior distribution with a mean of zero: ； in, It is a Gaussian density function with zero mean. yes The j-th element, and Represented by the gamma distribution: ； ； set up We can obtain this through Bayes' criterion. The posterior probability density function is: ； The mean and variance are: ； ； ； ； ； in, yes The j-th element, yes Obtain the j-th diagonal element and set the parameters appropriately. The above formula is recursively updated until the convergence criterion is met, and the mean is obtained. and variance The mean is used as the sparse reconstruction vector set. The estimated value can be obtained by effectively utilizing group sparsity through a multi-task Bayesian compressed sensing algorithm. To further obtain the target location The estimate.

5. The high-precision multi-station target detection method that integrates delay and frequency shift according to claim 4, characterized in that: Based on step four, the sparse reconstruction estimation vector is obtained. Then, the state subspace represented by the maximum value of the sum of sparse reconstruction vectors is taken as the estimated state, and the position represented by this subspace is taken as the estimated position. The mean distance difference between the estimated position and the actual position is calculated as the estimation error, and compared with the method of using only delay estimation.

Images