A method for optimizing a transmitting waveform against velocity deception jamming based on a quasi-orthogonal space

Abstract

The application relates to an optimization transmitting waveform anti-speed deception jamming method based on a quasi-orthogonal space and belongs to the technical field of radar signal processing, and is particularly suitable for the field where corresponding strong-power speed deception jamming exists in the main lobe or the side lobe of a radar under the condition of speed deception jamming. Firstly, radar intermediate frequency echo data are collected and down-converted to a baseband, are arranged into a radar echo data matrix after pulse compression, then a Fourier transform matrix is defined and a target to-be-estimated Doppler parameter interval is set; then a quasi-orthogonal space cost function is constructed, and a matrix phase and amplitude, a stop hyperparameter and an iteration number are initialized; finally, the phase value is updated according to an iteration formula, the final phase matrix and the transmitting waveform are updated after the iteration is stopped, and therefore, the speed deception jamming is effectively suppressed, so as to overcome the problems that in the prior art, a target signal-to-noise ratio is seriously lost and interference suppression capacity is insufficient.

Description

I. Technical Field

[0001] This invention belongs to the field of radar signal processing technology. II. Background Technology

[0002] With the increasing dimensionality of target parameters required for modern battlefield perception, accurately measuring and estimating the Doppler components of threatening targets to confirm their radial velocity has become a crucial aspect of battlefield information acquisition. Modern jammers, leveraging their one-way range advantage and employing radio frequency store-and-forward technology, can severely impact the Doppler detection of important targets. Traditional countermeasures against velocity deception jamming employ random phase methods or frequency agility, but these measures inevitably introduce a loss in the target's signal-to-noise ratio. Moreover, if the jamming power is strong enough to completely obscure the target signal, all these methods become ineffective or experience a significant performance degradation. Therefore, effectively suppressing velocity deception jamming with high power, while minimizing changes to the original radar information processing flow, and simultaneously reducing signal energy loss to improve radar target detection probability, is a critical technology in the field of radar signal processing that urgently needs to be addressed.

[0003] The University of Electronic Science and Technology of China (UESTC) disclosed an adaptive iterative estimation method for resisting velocity spoofing interference in its invention patent application, "An Adaptive Iterative Estimation Method for Resisting Velocity Spoofing Interference" (Publication No.: CN106249209A, Application No.: CN201610544322.1). This method mainly combines prior information from the transmitted signal, the interference signal, and the target signal. Then, by constructing optimal filters for the target and the interference at each Doppler unit, it ultimately suppresses velocity spoofing interference, effectively obtains the Doppler parameters of the target signal, and thus improves the target detection probability. However, this method uses an adaptive iterative approach, which is not suitable for situations with weak target signal-to-noise ratios. If the external noise power is high, it can lead to a significant increase in the number of iterations or even non-convergence, ultimately resulting in the failure of the anti-interference method and severely affecting the target detection probability. Therefore, this method lacks universality.

[0004] The paper "Anti-deception Interference Method Based on Joint Design of Inter-pulse Waveform Amplitude and Phase" (Journal of University of Electronic Science and Technology of China, 2021, Vol.50, pp:481-487) proposes an optimization problem for anti-velocity deception interference by "establishing an optimization criterion that minimizes the weighted sum of interference energy and target sidelobe energy within the stopband, while considering discrete quantization phase and peak-to-average power ratio (PAR) constraints." This method utilizes weighted optimization of interference energy within the stopband and target sidelobe energy to effectively suppress interference signals and achieve effective target detection. However, this algorithm does not utilize constant modulus constraints, and when introducing PAR constraints, it fails to consider the inaccuracy of parameter measurements caused by the noise floor, resulting in a huge computational load and potential for invalid results. Ultimately, this leads to a sharp decline in interference suppression performance, thus affecting normal target detection. III. Summary of the Invention

[0005] In the presence of high-power velocity deception jamming signals, traditional radar signal processing methods suffer from severe target signal-to-noise ratio loss and insufficient jamming suppression capability. This invention proposes an optimized transmit waveform anti-velocity deception jamming method based on quasi-orthogonal space.

[0006] To achieve the above technical objectives, the present invention adopts the following technical solution:

[0007] Step 1: Acquire radar echo data, down-convert to baseband, and construct the radar pulse-compressed echo data matrix after radar signal processing and pulse compression.

[0008] E(m,n)=T(m,n)+J(m,n)+ε

[0009] Where matrix E represents the pulse-compressed baseband radar echo data matrix, matrix T represents the matrix containing the target component, matrix J represents the matrix containing the velocity deception jamming component, and ε represents the complex Gaussian noise signal whose amplitude follows a Gaussian distribution. The radar pulse-compressed echo data matrix is ​​set to have a total of M range-gated data points and N velocity-gated data points; where m = 1, 2, 3…M represents the distribution unit of the pulse-compressed radar echo data matrix in the range dimension; and n = 1, 2, 3…N represents the distribution unit of the pulse-compressed radar echo data matrix in the Doppler dimension.

[0010] Step 2: After down-conversion to baseband, the pulse-compressed baseband radar echo data is obtained, mainly containing signal components and velocity deception jamming components. The influence of noise components is ignored here. The velocity deception jamming component can be described as follows: r = 1, 2, 3…L represents that a radar coherent pulse group has L pulses, n = 1, 2, 3…M represents the occupancy range gate for velocity deception jamming of each pulse, and the Doppler frequency estimation range of the target signal component is assumed to be... Where V1 and V2 represent the lower and upper bounds of the Doppler frequency estimation distribution of the target signal, respectively; construct the Fourier transform matrix D(q,w)=e j2πqw / K Where K represents the number of points in the Fourier transform of the radar target Doppler frequency estimation band, q and w are positive integers, with values ​​ranging from q = 1, 2, 3…K and w = 1, 2, 3…K; let D1 represent a new matrix composed of all columns selected from V1 to V2 in D, and D2 be a new matrix composed of the remaining columns. The problem of minimizing interference components in the target Doppler estimation region can be solved by constructing a quasi-orthogonal space cost function as follows:

[0011]

[0012] Here λ and S i The stopping hyperparameter is represented by i = 1, 2, 3…G, where G represents the maximum number of iterations. The cost function formula contains… This represents the constant modulus constraint for the velocity spoofing interference component, and ||·||2 represents the 2-norm of the matrix data, where the matrix in the cost function formula... Let represent the matrix composed of all velocity deception and interference components within a radar coherent pulse group in the i-th iteration. Here, it is assumed that a radar coherent pulse group has L pulses.

[0013] Step 3: After setting i=1, initialize the phase of the velocity deception interference matrix. This is a randomly distributed sequence matrix, where the superscript r = 1, 2, 3…L indicates that a radar coherent pulse group has L pulses, i = 1, 2, 3…G indicates the iteration number, with a maximum iteration number of G, and the amplitude is a fixed value. By combining each initialization element, we can obtain Simultaneously initialize the threshold comparison thresholds μ1 and μ2. Solve according to the formula. as well as The (·) here H This indicates that the Helmet transpose operation is performed, and arg() indicates that the phase calculation is performed.

[0014] Step 4: Comparison If the value is greater than the threshold μ1 or the number of iterations is greater than the threshold μ2, the loop exits, where abs(·) represents the trace operation of the matrix; otherwise, it returns to the previous step and continues iterating.

[0015] Step 5: Output the final transmitted waveform phase matrix sequence

[0016] This invention designs the waveform phase based on the quasi-orthogonal space of interference and signal. It constructs a quasi-orthogonal space cost function for actual data, and then iteratively optimizes and solves the problem to finally obtain the optimized transmitted waveform phase. This method has fast iterative convergence and is simple to implement. Compared with traditional methods that only consider the phase state of the transmitted waveform itself without considering interference data, this method has higher confidence and less signal-to-noise ratio loss for the target, thus having greater engineering application value.

[0017] The following is in conjunction with the appendix Figure 1 The present invention will be described in further detail below. IV. Description of the attached drawings

[0018] Figure 1 This is a preferred processing flowchart of the present invention.

[0019] Figure 2 This is a schematic diagram of iterative convergence in a specific embodiment of the present invention.

[0020] Figure 3 This is a Doppler detection image of the target after receiving velocity deception interference in a specific embodiment of the present invention.

[0021] Figure 4 This is a Doppler detection image after suppressing velocity deception interference using the method of the present invention in a specific embodiment of the present invention. V. Detailed Implementation Methods

[0022] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments, but this is not intended to limit the scope of the invention.

[0023] This invention proposes an optimized transmit waveform anti-velocity deception interference method based on quasi-orthogonal space, as illustrated in the schematic diagram below. Figure 1 As shown.

[0024] The feasibility of the method of the present invention will be further verified through experimental simulation below.

[0025] The experiment assumes that the radar uses a narrowband linear frequency modulated (LFM) signal for target detection. The radar is attacked by a high-power speed deception jamming signal. There are three target signals in the entire scene, all located at the same range cell. The normalized Doppler frequencies are 0.1Hz, 0.2Hz and 0.3Hz. The signal-to-noise ratios of the targets are set to 20dB, 10dB and -5dB. The radar pulse width is 1µs and the LFM signal bandwidth is 7MHz. One coherent processing contains 512 pulses. The radar pulse repetition frequency is set to 20kHz and the sampling rate is 10MHz. The threshold μ1 = 0.01, the iteration threshold μ2 = 18, and the jammer releases five speed deception jamming signals with signal-to-interference ratios (SIRs) (relative to the first target signal) of 20 dB, 20 dB, 20 dB, 10 dB, and 10 dB, respectively. The normalized Doppler frequencies of the jammer are 0.095 Hz, 0.105 Hz, 0.205 Hz, 0.21 Hz, and 0.31 Hz. Based on the output results before and after jamming suppression, the improvement in the SIR of the target signal and the convergence of the calculation are quantitatively analyzed, and the effectiveness of the method of this invention is finally analyzed.

[0026] Figure 2 The diagram shows the convergence number after iterative updates using the present invention. As can be seen from the convergence value of the penalty function and the number of iterations, convergence is basically achieved after 16 iterations, proving that the present invention can converge quickly.

[0027] Figure 3 The image shows three targets with different signal-to-interference-plus-noise ratios (SNRs) located in the same range cell, but with different normalized Doppler values. The first two targets have high SNRs, while the last target is completely submerged in interference.

[0028] Figure 4 The image shows the Doppler detection images of three targets after the spoofing interference was suppressed by the present invention. It can be seen that the signal-to-interference-plus-noise ratio (SNR) of all three targets is significantly improved, especially the SNR of the last target, which is improved by more than 45 dB. This proves that the present invention can effectively suppress high-power velocity spoofing interference and reduce the loss of target SNR.

[0029] This invention is not limited to this embodiment. Based on the technical solution disclosed in this invention, those skilled in the art can make some substitutions and modifications to some of the technical features without creative effort, and all such substitutions and modifications are within the protection scope of this invention.

Claims

1. A method for optimizing the transmitted waveform to resist velocity spoofing interference based on quasi-orthogonal space, characterized in that: a) Acquire radar echo data, down-convert to baseband, and construct a pulse-compressed baseband radar echo data matrix after radar signal processing and pulse compression: ； Where matrix E represents the baseband radar echo data matrix after pulse compression, matrix T represents the matrix containing the target component, and matrix J represents the matrix containing the velocity deception jamming component. ε The amplitude of the signal is a complex Gaussian noise signal that follows a Gaussian distribution. The radar echo data matrix after pulse compression is set to have a total of M range gate data and N velocity gate data; where m=1,2,3…M represents the distribution unit of the radar echo data matrix after pulse compression in the range dimension; n=1,2,3…N represents the distribution unit of the radar echo data matrix after pulse compression in the Doppler dimension. b) After down-conversion to baseband, the pulse-compressed baseband radar echo data is obtained, which includes signal components and velocity deception interference components, while ignoring the influence of noise components. The speed deception interference component is described as follows: Where r = 1, 2, 3… L represents that a radar coherent pulse group has L pulses, n = 1, 2, 3… M represents the number of range gates occupied by velocity deception jamming within each pulse, and the Doppler frequency estimation range of the target signal component is assumed to be... ,in V 1 and V 2 These represent the lower and upper bounds of the Doppler frequency estimation distribution of the target signal, respectively. Construct the Fourier transform matrix. Where K represents the number of points in the Fourier transform of the Doppler frequency estimation band of the radar target, q and w are positive integers, ranging from q=1,2,3…K, w=1,2,3…K; setting D1 indicates selecting points from D... V 1 to V 2 D2 is a new matrix composed of all columns between the two columns, and D2 is a new matrix composed of the remaining columns. The quasi-orthogonal space cost function is constructed to minimize the interference components in the target Doppler estimation region as follows: ； λ and S i This represents the stopping hyperparameter, where i = 1, 2, 3…G represents the number of iterations, with a maximum iteration count of G. The cost function formula contains… This represents the constant modulus constraint for the disturbance component. This represents calculating the 2-norm of matrix data, where the matrix in the cost function formula... Let represent the matrix composed of all interference components within a radar coherent pulse group in the i-th iteration, assuming that a radar coherent pulse group has L pulses; c) Initialize the phase of the interference matrix after setting i=1. This is a randomly distributed sequence matrix, where the superscript r = 1, 2, 3…L indicates that a radar coherent pulse group has L pulses, i = 1, 2, 3…G indicates the iteration number, with a maximum iteration number of G, and the amplitude is a fixed value. By combining each initialization element, we obtain Where i=1, and the threshold comparison threshold is initialized simultaneously. μ 1 and μ 2; d) Solve , The (●) here H This indicates that the Hermite transpose operation is performed, and arg() is used to calculate the phase. e) Calculation Compare when the MX value is greater than the threshold. μ 1 or the number of iterations is greater than the threshold μ 2. Exit the loop, where abs (●) indicates the trace operation of the matrix; otherwise, return to the previous step and continue the iteration. f) Output the final transmitted waveform phase matrix sequence .

Images