A multi-scale synchronous compression transform-based inverse synthetic aperture radar imaging method

Abstract

This invention discloses an inverse synthetic aperture radar (ISAR) imaging method based on multi-scale synchronous compression transform, which is achieved through the following steps: decomposing the target echo signal into low-frequency rigid body components and high-frequency micro-Doppler components; using coarse-scale synchronous compression to extract the low-frequency main structure, and fine-scale synchronous compression to separate high-frequency micro-motion features; using entropy normalization weights to fuse multi-scale time-frequency features to generate a high-resolution fused spectrum; reconstructing the target image based on the fused spectrum, simultaneously suppressing noise and enhancing micro-motion features. This invention, through multi-scale hierarchical processing and entropy weight fusion mechanism, solves the problems of image blurring and micro-Doppler aliasing caused by acceleration motion, significantly improving the imaging clarity and feature recognition capability of high-speed maneuvering targets in complex scenes, and providing a reliable solution for high-resolution radar imaging.

Description

Technical Field

[0001] This invention relates to the field of inverse synthetic aperture radar (ISAR) imaging technology, and in particular to a high-resolution ISAR imaging method based on multi-scale synchronous compression transform (MSST). Background Technology

[0002] Inverse Synthetic Aperture Radar (ISAR) is an important target imaging technology widely used in military reconnaissance, target identification, and space surveillance. However, in complex scenarios, traditional ISAR imaging methods face problems such as insufficient resolution and severe noise interference. In recent years, time-frequency analysis techniques have been introduced into ISAR imaging, but the time-frequency distribution at a single scale is insufficient to simultaneously capture the rigid structure and micro-motion features of the target. To address this, this invention proposes an ISAR imaging method based on Multi-Scale Synchronous Compression Transform (MSST), which significantly improves imaging resolution and noise resistance through multi-resolution analysis and adaptive synchronous compression technology. Summary of the Invention

[0003] The purpose of this invention is to provide an ISAR imaging method based on multi-scale synchronous compression transform (MSST), which can effectively extract the rigid structure and micro-motion features of the target and generate high-resolution ISAR images.

[0004] To achieve the above objectives, the present invention adopts the following technical solution: an inverse synthetic aperture radar imaging method based on multi-scale synchronous compression transform, the method comprising:

[0005] S1. Input the echo signal and perform preprocessing;

[0006] S2. Decompose the signal using multi-resolution analysis techniques and perform synchronous compression transformation;

[0007] S3. Extract multi-scale features and calculate weights using normalized entropy, then fuse them to generate an optimized time-frequency distribution;

[0008] S4. Generate high-resolution ISAR images based on the fused time-frequency distribution;

[0009] The technical advantages of this invention are as follows: it improves the time-frequency resolution through multi-scale feature extraction and fusion; it optimizes the weight allocation using normalized entropy to ensure the robustness of feature extraction; and experimental verification shows that the method performs excellently in imaging. Attached Figure Description

[0010] Figure 1 This is a flowchart illustrating the inverse synthetic aperture radar imaging method based on multi-norm constraints.

[0011] Figure 2 This is the target model diagram.

[0012] Figure 3 This is the imaging result image of the SST algorithm.

[0013] Figure 4 The image shows the imaging results of the SST algorithm with entropy optimization.

[0014] Figure 5 This is an image showing the imaging results from the MSST algorithm. Detailed Implementation

[0015] The specific embodiments of the present invention are described below to enable those skilled in the art to understand the present invention. However, it should be understood that the present invention is not limited to the scope of the specific embodiments. For those skilled in the art, various changes are obvious as long as they are within the spirit and scope of the present invention as defined and determined by the appended claims. All inventions utilizing the concept of the present invention are protected.

[0016] like Figure 1 As shown, an inverse synthetic aperture radar imaging method based on multi-scale synchronous compression transform includes the following steps:

[0017] S1. Receive the original echo signal x(t) from the target and preprocess the signal to remove noise and interference.

[0018] S2. Using multi-resolution analysis techniques, the signal is decomposed into multiple scale components, and a synchronous compression transformation is performed on each scale component to generate a high-resolution time-frequency distribution. The formula for calculating the time-frequency distribution is:

[0019]

[0020] W(i,k,f) represents the time-frequency representation at scale k and frequency f, i represents the time index of the signal, S(i,k) represents the signal decomposition coefficients at time i and scale k, ψ(t) represents the wavelet basis function, and the superscript * indicates conjugate.

[0021] The energy is then concentrated into a more precise time-frequency region using a synchronous compression transformation formula:

[0022]

[0023] Where, ω i,k This represents the center frequency corresponding to time index i and scale k, and δ(.) represents the Dirac delta function, used for precise positioning in the frequency domain.

[0024] S3: Multi-scale Feature Extraction and Fusion

[0025] Rigid body structural features and micro-motion features are extracted from each scale component. Weights are calculated using normalized entropy, and the multi-scale features are weighted and fused to generate an optimized time-frequency distribution. The formula for calculating normalized entropy is:

[0026]

[0027] Among them, H(SST) s (t,f) represents SST s(t,f) The Shannon entropy, ∈, represents a small positive constant, and the probability distribution P... s (t) is defined as:

[0028]

[0029] The formula for the fusion spectrum is:

[0030]

[0031] Among them, w s As weight, SST s (t,f) represents the synchronous compression transformation result at the s-th scale, SST fused (t,f) represents the final synchronous compression transformation result after multi-scale fusion.

[0032] S4: Image Reconstruction and Output

[0033] Based on the fused time-frequency distribution, a high-resolution ISAR image is reconstructed using inverse transform technology. The inverse transform formula is:

[0034]

[0035] Where I(x,y) represents the generated two-dimensional ISAR image, This represents the inverse Fourier transform.

[0036] In one embodiment of the invention, an experiment simulated an accelerating moving target scenario using MATLAB. The radar operated at a center frequency of 31 GHz and a bandwidth of 6 GHz. The target consisted of a rigid body and micro-motion scattering points. The experiment used UAV scattering point model data, setting the target with an initial velocity of 1.5 m / s and a scattering point of 0.1 m / s². 2 The acceleration is in the horizontal direction. First, the distance from each scattering point to the radar and its change are calculated, generating the corresponding radar echo signal. For example... Figure 2 As shown, the echo profile before distance alignment is displayed.

[0037] The basic SST method first preprocesses the echo signal using the Kaiser window function, then analyzes the signal at multiple scales using adaptive Morlet wavelets, and finally synthesizes the results for two-dimensional FFT imaging. Figure 3As shown, although the imaging results of the SST method show the basic outline of the target, the details are blurred, the edges of the scattering center are not clear, there is a lot of background noise, and it is difficult to accurately identify the small structural features of the target.

[0038] The SST method with entropy optimization adds entropy optimization and signal enhancement steps to the basic algorithm. It introduces a Gaussian weighting function to weight the wavelet transform results and applies an entropy optimization factor to adjust the signal energy distribution. For example... Figure 4 As shown, the imaging results of this method are compared to Figure 3 There is a significant improvement; the target outline is clearer, and background noise is partially suppressed. However, the ability to resolve scattering points in complex areas remains limited.

[0039] The MSST method of this invention achieves adaptive fusion of multi-scale information by expanding the wavelet scale range to 512 scales and calculating weights based on inverse variance for each scale. For example... Figure 5 As shown, the image generated by the MSST method has the clearest target scattering center distribution, the sharpest boundaries, and the most effective suppression of background noise. Even in areas with complex target structures, each scattering point can be clearly distinguished, demonstrating the superior performance of this method in processing images of accelerating moving targets.

[0040] Table 1 compares the results of different algorithms under the same simulation conditions.

[0041] algorithm entropy ECI SNR SST 0.08 0.69 0.19 SST with entropy optimization 1.29 0.30 1.95 MSST 1.60 0.19 2.09

[0042] As shown in Table 1, SST achieved lower entropy (0.08) and SNR (0.19), indicating weaker energy concentration and significant noise. SST with entropy optimization increased the entropy to 1.29, the SNR to 1.95, and reduced the ECI to 0.30. MSST achieved the highest entropy (1.60) and SNR (2.09), with a slight reduction in ECI to 0.19, demonstrating its effectiveness in improving energy concentration and reducing noise. The MSST method outperforms traditional methods in terms of imaging resolution and noise suppression.

Claims

1. An inverse synthetic aperture radar imaging method based on multi-scale synchronous compression transform, characterized in that: Includes the following steps: S1. Receive the original echo signal from the target and preprocess the signal to remove noise and interference; S2. Decompose and synchronously compress the signal, separating low-frequency and high-frequency components and performing adaptive compression. In the signal decomposition and synchronous compression process described in S2, the time-frequency distribution is first calculated. The formula is: ； , representing the scale and frequency The following time-frequency representation, Indicates the time index of the signal. Indicates time index and scale The signal decomposition coefficients below, Denotes wavelet basis functions, superscript Indicates conjugate; The energy is then concentrated into a more precise time-frequency region using a synchronous compression transformation formula: ； in, Indicates time index and scale The corresponding center frequency, This represents the Dirac function, used for precise positioning in the frequency domain; S3. Extract rigid body structure and micro-motion features from the signal, and calculate the weights by normalized entropy. Then, weighted fuse the multi-scale features to generate an optimized time-frequency distribution. The formula for calculating entropy in step S3 is as follows: ； in, express Shannon entropy, Given a positive constant; probability distribution Defined as: ； The formula for the fusion spectrum is: ； in, As weight, This represents the result of the synchronous compression transformation at the s-th scale. This represents the final synchronous compression transformation result after multi-scale fusion; S4. Generate high-resolution ISAR images based on the fused time-frequency distribution.

2. The inverse synthetic aperture radar imaging method based on multi-scale synchronous compression transform according to claim 1, characterized in that, In step S4, the high-resolution ISAR image is reconstructed using inverse transform technology. The inverse transform formula is: ； in, This represents the generated two-dimensional ISAR image. This represents the inverse Fourier transform.

3. A method for inverse synthetic aperture radar imaging based on multi-scale synchronous compression transform according to any one of claims 1 to 2, characterized in that, The method was experimentally verified using a radar system with a center frequency of 30 GHz and a bandwidth of 1 GHz. The evaluation metrics included entropy, energy concentration, and signal-to-noise ratio.

Images