Shuwen Xu

Range-Spread Target Detection using Consecutive HRRPs

In this paper, a heuristic detector is proposed to detect range-spread targets in white Gaussian noise using multiple consecutive high- resolution range profiles (HRRPs) received from a high-resolution radar (HRR). The detector consists of refiners of HRRPs and a cross-correlation integrator of refined HRRPs. Based on the fact that strong scattering cells are sparse in target HRRPs, nonlinear shrinkage maps are designed to refine received HRRPs before integration, by which most of the noise-only cells in received HRRPs are suppressed while strong scattering cells most probably relevant to target signature are preserved. Since the targets scattering geometry is almost unchanged except for range walking during integration, the refined target HRRPs from consecutive pulses are highly similar while refined noise-only HRRPs are dissimilar due to randomicity. The modified correlation matrix of multiple refined HRRPs is used to measure their similarity. The test statistic, a weighted integration of the entries of the modified correlation matrix, is constructed for target detection. The proposed detector does not depend on a strict target return model and can work in mild conditions. The real target data and simulated noise are used to evaluate the detector, and the experimental results show that it achieves better detection performance than some existing methods.

Adaptive range-spread maneuvering target detection in compound-Gaussian clutter

Adaptive detection of range spread maneuvering target embedded in compound-Gaussian clutter is an important challenge for radar engineers. For the long integration, maneuvering target suffers from inevitable range walks across cells as well as unpredictable phase change at individual range cells. Therefore, the traditional adaptive normalized matched filter detectors are ineffective in long integration duration. In this paper, combining adaptive normalized matched filter with sliding high-order cross-correlation integration, a new adaptive range-spread target detector is proposed. Firstly, the long integration duration is segmented into short disjoint subintervals. In each subinterval, it is assumed that no range walking across cells happens and targets complex returns accord with the traditional rank-one parameter models. In each subinterval, the coherent integration output vector is obtained by the adaptive normalized matched filter along the slow-time dimension in different range cells. If a target is present, the coherent integration outputs of each subinterval share similar waveforms except for unknown range walks. Otherwise, when a target is absent, the coherent integration output vectors estimated from individual subintervals are uncorrelated positive random vectors without similarity. Further, the sliding high-order cross-correlation integration of these coherent integration output vectors is calculated for target detection. The new detector combines the short-time coherent integration and long-time similarity integration. The experimental results using raw radar target data and simulated clutter data show that the new detector achieves better detection performance for range-spread maneuvering targets than the existing detectors.

Tri-feature-based detection of floating small targets in sea clutter

It is always a challenging problem for marine surface surveillance radar to detect sea-surface floating small targets. Conventional detectors using incoherent integration and adaptive clutter suppression have low detection probabilities for such targets with weak returns and unobservable Doppler shifts. In this paper, three features of a received vector at a resolution cell-the relative amplitude, relative Doppler peak height, and relative entropy of the Doppler amplitude spectrum-are exploited to give returns with targets from sea clutter. Real datasets show that each feature alone has some discriminability, and the three features jointly exhibit strong discriminability. Due to diversity of targets in practice, it is impossible to get features of returns with all kinds of targets. We recast detection of sea-surface floating small targets as a one-class anomaly detection problem in the 3D feature space. A fast convexhull learning algorithm is proposed to learn the decision region of the clutter pattern from feature vectors of clutter-only observations. As a result, a tri-feature-based detector is developed. The experiment results for the IPIX datasets show that the proposed detector at an observation time of several seconds attains better detection performance than several existing detectors.

SAR image segmentation based on Gabor filters of adaptive window in overcomplete brushlet domain

In this paper, a new technique based on Gabor filters with adaptive window is proposed for SAR image segmentation in overcomplete brushlet domain. SAR image is full of texture and direction information, and brushlet is a new kind of analysis tool for image with rich directional information. Aim at these characteristics, this paper combines Gabor filters with GLCP for segmentation in brushlet coefficients domain. In SAR image processing, the difficulty is the balance between regions and boundaries. The adaptive window technique is adapted in this paper to obtain the optimal size of Gabor filters. The presented method was found to be superior to three other methods by the experimental results for SAR image.

Shape-parameter-dependent coherent radar target detection in K-distributed clutter

In this paper, shape-parameter-dependent matched filter (MF) detectors are proposed for moving target detection in K-distributed clutter, which are specified by a single parameter α ∈ [0, 1], the α-MF detectors for short. The α-MF detectors include the MF and normalized matched filter (NMF) detectors as special examples with α = 0 and 1. The parameter α can be chosen to match clutter characteristics. An empirical formula is given that the optimal parameter α equals the number of integrated pulses divided by it plus the shape parameter of K-distributed clutter. It is proved that the α-MF detectors are constant false alarm rate (CFAR) with respect to the scale parameter, clutter covariance matrix, and Doppler steering vector. The properties of adaptive α-MF (α-AMF) detectors are discussed. For K-distributed clutter, the optimal α-MF detectors are superior to the MF and NMF detectors and are comparable with the optimal K-distributed detectors (OKDs) of high computational cost. Finally, real high-resolution sea clutter data are available to verify the proposed detectors. The optimal α-AMF detectors under K-distributed clutter model are competitive in performance with the adaptive generalized likelihood ratio test linear threshold detectors (GLRT-LTDs) that are optimal for the compound-Gaussian clutter with the inverse Gamma texture.

Fast communication: Range-spread target detection using 2D non-local nonlinear shrinkage map

A novel 2D non-local nonlinear shrinkage map (NLNSM) on 2D range-pulse image is proposed in this paper and it is imbedded into the geometric average integration (GAI) scheme to yield a new range-spread target detector in white Gaussian noise. Because the denoising ability of the NLNSM is better than that of the 2D local nonlinear shrinkage map (NSM), the new detector attains better detection performance than the previous GAI detector, which is testified by measured target data.

Adaptive detection of range-spread targets in compound Gaussian clutter with the square root of inverse Gaussian texture

The compound Gaussian clutter with the square root of inverse Gaussian texture component has been successfully used for modeling the heavy-tailed non-Gaussian clutter measured by high-resolution radars. In high-resolution radars, the targets may extend along multiple consecutive range cells, which are called range-spread targets. In this paper, we consider the range-spread target detection problem in the compound Gaussian clutter with the square root of inverse Gaussian texture. Three adaptive detectors are proposed based on Bayesian one-step generalized likelihood ratio test, maximum a posteriori generalized likelihood ratio test and Bayesian two-step generalized likelihood ratio test, respectively. Finally, the detection performances of the proposed detectors are evaluated by the Monte Carlo simulation. The simulation results show that the proposed detectors have better detection performance of range-spread target than the conventional generalized likelihood ratio test detector.

Double-characters detection of nonlinear frequency modulated signals based on FRFT

In many practical applications, signals to be detected are unknown nonlinear frequency modulated (FM) and are corrupted by strong noise. The phase histories of the nonlinear FM signals are assumed to be unknown smooth functions of time, which are usually poorly modeled or cannot be modeled at all by a small number of parameters. Because of the lack of phase model, a nonparametric detection method is proposed based on successive fractional Fourier transform and double-characters detection. The detection process goes in three steps. First, an image is constructed by the fractional Fourier transforms of successive angles in one period. Then, the threshold procedure is utilized to transform the image into a binary image. After the multiple median filtering, the binary image is refined where the isolated noise pixels are removed. Finally, two complementary features are extracted from the refined image, and a double-characters detector is proposed to decide whether the target is present or not. The simulation experiments to three polynomial phase signals with different orders and a sinusoidal phase signal show that the proposed detection method is effective and robust.

Knowledge-based adaptive detection of radar targets in generalized Pareto clutter

Abstract This paper studies adaptive detection of radar targets embedded in generalized Pareto clutter on the condition with the limited secondary data. In order to alleviate the effects of the non-Gaussian characteristic of the clutter, a-priori knowledge of the non-Gaussian clutter is considered in the designed detector. More precisely, we consider that the texture of clutter obeys the inverse gamma distribution and the inverse covariance matrix of speckle is a combination of multiple a-priori spectral models. Within these considerations, we obtain an adaptive detector based on the generalized likelihood ratio test. Finally, the performance of the proposed detector is evaluated via the Monte-Carlo technique. The experiments results indicate that the proposed detector outperforms the 1S-GLRT detector in limited secondary data scenarios.

Combined Adaptive Normalized Matched Filter Detection of Moving Target in Sea Clutter

In this paper, combined adaptive normalized matched filter (ANMF) detector is proposed to detect moving target in sea clutter. For the long integration time, moving target suffers from Doppler frequency changes in individual range cells, and the sea clutter is nonstationary along the pulse dimension. Therefore, ANMF detector is ineffective in this situation. In order to solve this problem, combined ANMF detector is proposed to detect moving target in sea clutter. Firstly, the long integration duration is segmented into some short subintervals. In each subinterval, the normalized Doppler frequency of the moving target is assumed to be a constant. Secondly, a series of ANMF detectors with different normalized Doppler frequencies are used to obtain a train of coherent integration results in each subinterval, which constitute the ANMF output vector. Finally, we use the multiply integration (MI) to integrate the ANMFOVs from different subintervals. The test statistic, the max-value of MI along the normalized Doppler frequency dimension, is used for the proposed combined ANMF detector. The real sea clutter and simulated clutter data are used to evaluate the proposed detector, and the experimental results show that the proposed detector achieves better detection performance than the conventional ANMF detector.