Inhomogeneous clutter suppression method based on Mahalanobis distance reweighting

Abstract

The invention belongs to the technical field of radar signal processing and discloses an inhomogeneous clutter suppression method based on Mahalanobis distance reweighting. The method comprises steps of: determining eliminating coefficients corresponding to respective pieces of sampled data according to the Mahalanobis distances between the sampled data and reference data; eliminating interference samples from sampling samples by using the eliminating coefficient to obtain data samples; then selecting corresponding training samples according to the Mahalanobis distances between the data samples and the reference samples, and then constructing a reweighting coefficient for reweighting by further using the Mahalanobis distances between the training samples and the reference samples to obtain a corresponding covariance matrix; finally, constructing an adaptive weight vector by using the covariance matrix, and performing clutter suppression by using the adaptive weight vector to obtain corresponding output power. The method can obtain good clutter suppression performance in the case of different sampling number, and has universality and robustness to the clutter inhomogeneous suppression in complex clutter background.

Description

technical field [0001] The invention belongs to the technical field of radar signal processing, in particular to a non-uniform clutter suppression method based on Mahalanobis distance reweighting, which is used for clutter suppression and slow target detection in a non-uniform clutter environment. Background technique [0002] Because the airborne early warning aircraft radar is often in the working state of looking down, its working performance will inevitably be strongly affected by ground clutter. At the same time, due to the relative movement between the airborne radar platform and the ground, the clutter received by the airborne radar presents a strong space-time coupling, which brings great difficulties to the detection of weak and slow targets. In the 1970s, Brennan et al. proposed the Space-Time Adaptive Processor (STAP) method, which suppresses clutter to achieve weak and slow targets through two-dimensional processing in the air and time domains. purpose of detect...

AI Technical Summary

Problems solved by technology

Existing NHDs include: (1) Generalized Inner Product (GIP), GIP has a better detection effect when there is an obvious target signal in the training sample, but since the generalized inner product method only extracts the Large eigenvalues, so in the case of multiple target signals, its detection performance will decline; Adaptive Power Residue (APR) detector, also the detector has a small number of target signals in the training samples, and the signal-to-noise A better detection effect can be obtained when the ratio is relatively low. When the number of target signals in the training samples is large and the signal-to-noise ratio is high, the interference-plus-noise covariance matrix formed by the training samples will be severely disturbed by the target signal, and the APR detection The performance of the detector is degraded; the diagonally loaded generalized inner product detector (Ioaded Generalized Inner Product, LGIP), which reduces the disturbance of the target signal to the detector by diagonally loading the covariance matrix estimated from the training samples, has How to choose the appropriate loading coefficient to improve the detection performance; the generalized inner product detector (Prolate SpheroidalWave Functions GIP, PSWF-GIP) based on the prolate The clutter covariance matrix of the detection sample avoids the influence of the target signal in the training sample, and can achieve better performance in theory. However, when calculating the basis vector of the prolate spherical wave function, the radar configuration parameters and the array flow pattern are required to be accurate. Known, but the array error and other factors in the actual situation will affect the array flow pattern and radar configuration parameters, making the detection performance degrade

Images