A Fast Solution Implementation Method Based on Iterative Space-Time Adaptive Processing

Abstract

This invention belongs to the field of signal and information processing technology, and relates to a fast solution implementation method based on iterative space-time adaptive processing, including the following steps: Step 1: The radar array receives echo data; Step 2: Estimate the detection range gate clutter plus noise covariance matrix based on the echo data received by the radar array; Step 3: Output the target steering vector, preprocess the covariance matrix and initialize the weight vector w. k Step 4: Set the iteration number K according to the clutter influence of the detection area; Step 5: Start iterative calculation of the weight vector and update the weight vector; Step 6: If k≤K, then k=k+1, return to step 5, otherwise proceed to step 6; Step 7: Set the final w k Perform normalization to obtain the processor weight vector w; Step 8: Perform target detection for each distance gate based on the processor weight vector.

Description

Technical Field

[0001] This invention belongs to the field of signal and information processing technology, and relates to a fast solution implementation method based on iterative spatiotemporal adaptive processing. Background Technology

[0002] In the field of radar target detection, the STAP (Spatial-Time Adaptive Processing) technique, which provides rapid solutions, has attracted widespread attention due to its ability to effectively suppress clutter. STAP fully utilizes multiple spatial domain information provided by multi-channel radar and temporal domain information provided by coherent pulses. Through two-dimensional adaptive joint filtering in both spatial and temporal domains, it can effectively suppress the influence of clutter in target detection. However, conventional solution methods involve high-dimensional matrix inversion operations, resulting in high computational complexity that prevents real-time implementation, thus becoming one of the main obstacles to the engineering implementation of STAP.

[0003] Existing fast STAP implementation methods mainly employ dimensionality reduction techniques, such as the mDT or JDL methods, to reduce the matrix dimension and thus computational complexity. However, these methods sacrifice some clutter suppression performance and still involve direct matrix inversion operations, lacking flexibility and thus performing poorly in practical engineering applications. Summary of the Invention

[0004] The purpose of this invention is to avoid the high computational complexity caused by high-dimensional matrix inversion in traditional STAP algorithms, and to address the lack of flexibility in existing dimensionality-reduction STAP methods, by providing an iterative-based fast STAP implementation method. This invention obtains the high-dimensional matrix using conventional methods, then treats the STAP filter weight vector solution problem as an unconstrained quadratic convex optimization problem. It directly iteratively solves the weight vector using the preprocessed conjugate gradient method. When the clutter influence in the detection region is small, a small number of iterations can be set; when the clutter influence is large, the number of iterations can be increased accordingly, allowing for flexible trade-offs between clutter suppression performance and computational time. This invention not only reduces the computational burden of traditional STAP methods but can also be extended to other STAP processing methods that require high-dimensional matrix inversion, achieving fast and flexible clutter suppression.

[0005] The technical solution of the present invention:

[0006] A fast solution implementation method based on iterative space-time adaptive processing includes the following steps:

[0007] Step 1: The radar array receives echo data;

[0008] Step 2: Estimate the detection range gate clutter plus noise covariance matrix based on the echo data received by the radar array;

[0009] Step 3: Draw the target guidance vector, preprocess the covariance matrix and initialize the weight vector w. k k = 0;

[0010] Step 4: Set the number of iterations K based on the degree of clutter influence in the detection area;

[0011] Step 5: Begin iteratively calculating and updating the weight vector;

[0012] Step Six: If k ≤ K, then k = k + 1 and return to Step Five; otherwise, proceed to Step Six.

[0013] Step 7: Place the last w k Perform normalization to obtain the processor weight vector w;

[0014] Step 8: Perform target detection for each distance gate based on the processor weight vector.

[0015] Furthermore, in step one, the echo data for the l-th range gate that does not contain the target is...

[0016]

[0017] Where, σ i Let N be the random complex clutter echo intensity received by the i-th clutter scattering unit, which follows a normal distribution; n is Gaussian white noise, and N is... c Indicates the number of clutter scattering units, v(f si ,f di Let ) be the space-time steering vector of the i-th clutter scattering unit, expressed as:

[0018]

[0019] in This represents the Kronecker product, where N is the number of radar elements, M is the number of pulses received by each radar element, and f si and f di Let be the spatial frequency and Doppler frequency of the i-th clutter scattering unit, respectively, and be expressed as:

[0020]

[0021] Where θ i and Here, λ represents the azimuth and elevation angles of the clutter scattering element relative to the radar array, v represents the velocity of the clutter scattering element relative to the radar array, and f represents the velocity of the clutter scattering element relative to the radar array. PRF is the pulse repetition frequency.

[0022] Furthermore, in step two, the data from the two range gates before and after the detection range gate are treated as protection units. A total of L = 2NM sample data points, excluding the target range gate and the protection units, are selected to estimate the noise covariance matrix R. The formula is as follows:

[0023]

[0024] The superscript H represents the conjugate transpose.

[0025] Furthermore, in step three, the target guidance vector can be represented as:

[0026]

[0027] Where f st and f dt Let be the target's spatial frequency and Doppler frequency, respectively, denoted as:

[0028]

[0029] where and θ t and These are the azimuth and elevation angles of the target relative to the radar array, respectively. t The velocity of the target relative to the radar array

[0030] The preprocessing operation involves normalizing all other data using the data from the first row and first column of the covariance matrix, as shown in the following formula:

[0031] R′=R. / abs[R(1,1)]

[0032] R′ is the preprocessed covariance matrix;

[0033] The initial weight vector is represented as follows: w0 = ones(NM,1).

[0034] Furthermore, in step five, the iterative calculation process is as follows:

[0035] Calculate the current gradient:

[0036]

[0037] Calculate the current conjugate gradient:

[0038]

[0039] in,

[0040] Calculate the iteration step size:

[0041]

[0042] Furthermore, in step five, the formula for updating the weight vector is as follows:

[0043] w k =w k-1 +α k p k .

[0044] Furthermore, in step seven, the formula for normalizing the weight vector is as follows:

[0045]

[0046] Furthermore, in step eight, the target detection formula for each distance gate based on the processor weight vector is as follows:

[0047] y l =w H x l .

[0048] The beneficial effects of this invention are:

[0049] This invention designs an iterative space-time adaptive processing fast solution implementation method. This method can converge quickly, solves the problem of high computational complexity of direct inversion of high-dimensional matrices in traditional STAP processing, and can adjust the number of iterations according to the clutter influence in the detection area. It has the characteristics of flexible interchangeability between clutter suppression performance and computation time. Attached Figure Description

[0050] Figure 1 A flowchart of a fast solution implementation method based on iterative space-time adaptive processing;

[0051] Figure 2 This is a simulation diagram showing the convergence of the method of the present invention with the number of iterations;

[0052] Figure 3 Simulation plots of the improvement factors for different iteration numbers;

[0053] Figure 4 Simulation output images of target detection for different iteration numbers;

[0054] Figure 5 This is a comparison chart of the computational complexity of the method of this invention and the direct inversion method. Detailed Implementation

[0055] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, 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 of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0056] like Figure 1 As shown, the process of this invention is as follows: First, the clutter covariance matrix of the detection area sample is calculated and preprocessed based on the echo data received by the radar system. Iteration parameters are initialized, and the number of iterations is set according to the influence of the detection area. Finally, the normalized STAP filter weight vector is obtained through iterative calculation. Specifically, the two range gate data before and after the detection range gate are treated as protection units. A certain number of range gate sample data are used to estimate the clutter covariance matrix of the detection area, and the matrix is ​​preprocessed. An initial weight vector and initial iteration parameters are set, and the number of iterations is set according to the magnitude of the clutter influence in the detection area. If the clutter influence is small, a small number of iterations is set; if the clutter influence is large, the number of iterations is increased. Then, iterative calculation is performed until the iteration ends, and finally, the normalized STAP filter weight vector is obtained.

[0057] Step 1: The radar array receives echo data;

[0058] The radar echo data for the l-th range gate that does not contain the target is,

[0059]

[0060] Where σ i To satisfy a certain distribution of random complex clutter echo intensity, we assume it follows a log-normal distribution. n is Gaussian white noise, and v(f si ,f di The space-time steering vector of the i-th clutter scattering unit can be expressed as:

[0061]

[0062] in f represents the Kronecker product. si and f di The spatial frequency and Doppler frequency, respectively, can be expressed as:

[0063]

[0064] Where θ i and These represent the azimuth and elevation angles of the clutter scattering element relative to the radar array, respectively.

[0065] Step 2: Estimate the detection range gate clutter plus noise covariance matrix based on the radar received data;

[0066] Using the range gate data before and after the detection range gate as protection units, a total of L = 2NM sample data points (excluding the target range gate and protection units) are selected to estimate the clutter covariance matrix.

[0067]

[0068] Step 3: Given the target guidance vector, preprocess the noise covariance matrix and initialize the iteration parameters.

[0069] The target guidance vector can be represented as:

[0070]

[0071] Preprocessing covariance matrix

[0072] R′=R. / abs[R(1,1)]

[0073] Set the initial STAP filter weight vector as follows:

[0074] w0 = ones(NM, 1)

[0075] Step 4: Set the number of iterations K according to the degree of clutter influence in the detection area, and start iteratively calculating the weight vector.

[0076] The specific iterative steps are as follows:

[0077] Find the current gradient.

[0078]

[0079] Find the current conjugate gradient.

[0080]

[0081] Where β k The solution formula is as follows:

[0082]

[0083] Calculate the iteration step size.

[0084]

[0085] Step 5: Update the STAP weight vector.

[0086] w k =w k-1 +α k p k

[0087] Step 6: Repeat steps 4 and 5 until the number of iterations is K.

[0088] Step 7: Put the last w k After normalization, the STAP processor weight vector w is obtained.

[0089]

[0090] Figure 2The results show the convergence of the method under different iteration numbers. It can be seen that the clutter suppression performance basically converges when the number of iterations is greater than 10. This demonstrates that the method has a fast convergence speed and can effectively reduce computational complexity. Figure 3 By setting the number of iterations to 5, 10, and 50, the corresponding improvement factor diagrams were obtained, further demonstrating that the proposed method can converge quickly and achieve ideal clutter suppression performance with a small number of iterations.

[0091] Step 8: Perform STAP target detection for each distance gate:

[0092] y l =w H x l

[0093] Example:

[0094] Assume an airborne radar system is a uniform linear array with N = 32 channels, element spacing of half a wavelength, transmitting M = 16 pulses within one coherent pulse processing interval, operating in a front-side-looking configuration. The radar operating frequency is 450 MHz, pulse repetition frequency is 1200 Hz, and signal bandwidth is 5 MHz. The airborne radar platform altitude is H = 12 km, and platform speed is V = 120 m / s. Assume the ambient clutter-to-noise ratio (CNR) is 40 dB, the target speed is 20 m / s, located 50 km from the radar system (i.e., the 480th range gate), and the signal-to-clutter ratio (SCR) is 10 dB. Clutter scattering elements are modeled as N equal intervals at azimuth angles from 0 to 180 degrees on each range gate. c = 180 scattering points.

[0095] Figure 4 The results show the STAP target detection power output corresponding to 5, 10, and 50 iterations. It can be seen that if the clutter effect in the detection area is small, a small number of iterations is sufficient to achieve ideal clutter suppression performance. In this experiment, setting the number of iterations to 5 was sufficient to detect the target effectively.

[0096] Figure 5 The comparison between this method and direct inversion shows that this method can significantly reduce the computational cost, and the effect of reducing the computational cost is more obvious when the matrix dimension is higher.

[0097] The above description is merely a specific embodiment of the present invention, providing a detailed description of the invention. Parts not covered herein are conventional techniques. However, the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention. The scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A fast solution implementation method based on iterative space-time adaptive processing, characterized in that: Includes the following steps: Step 1: The radar array receives echo data; Step 2: Estimate the detection range gate clutter plus noise covariance matrix based on the echo data received by the radar array; Step 3: Provide the target guidance vector ,in and Let be the target's spatial frequency and Doppler frequency, respectively, denoted as: Among them and and These are the azimuth and elevation angles of the target relative to the radar array, respectively. The velocity of the target relative to the radar array; Preprocess the covariance matrix and initialize the weight vector ; Step 4: Set the number of iterations based on the degree of clutter influence in the detection area. K ; Step 5: Begin iteratively calculating and updating the weight vector, as follows: Calculate the current gradient: Calculate the current conjugate gradient: in, Calculate the iteration step size: ; The formula for updating the weight vector is as follows: ; Step 6: If k ≤ K, then k = k + 1 and return to step 5; otherwise, proceed to step 6. Step 7: Place the last Perform normalization to obtain the processor weight vector. ; Step 8: Perform target detection for each distance gate based on the processor weight vector.

2. The method according to claim 1, characterized in that: In step one, the echo data of the first... l The echo data of the distance gate that does not contain the target is: in, For the first i The intensity of the random complex clutter echo received by each clutter scattering unit follows a normal distribution. n It is Gaussian white noise. N c Indicates the number of clutter scattering units. For the first i The space-time steering vector of each clutter scattering unit is expressed as: in Represents the Kronecker product. N This refers to the number of radar array elements. M The number of pulses received for each radar element. and The first i The spatial frequency and Doppler frequency of each clutter scattering unit are expressed as: in and Here, λ represents the azimuth and elevation angles of the clutter scattering element relative to the radar array, respectively, and λ is the wavelength. v The velocity of the clutter scattering element relative to the radar array. is the pulse repetition frequency.

3. The method according to claim 2, characterized in that: In step two, the data from the two range gates before and after the detection range gate are used as protection units, and common range gates other than the target range gate and protection units are selected. L =2 NM Estimate the noise covariance matrix from each sample data. R The formula is as follows: The superscript H represents the conjugate transpose.

4. The method according to claim 3, characterized in that: In step three, The preprocessing operation involves normalizing all other data using the data from the first row and first column of the covariance matrix, as shown in the following formula: This is the preprocessed covariance matrix.

5. The method according to claim 4, characterized in that: In step seven, the formula for normalizing the weight vector is as follows: 。 6. The method according to claim 5, characterized in that: In step eight, the target detection formula for each distance gate based on the processor weight vector is as follows: 。

Images