基于稀疏恢复的直接数据域STAP算法

在机载/星载雷达系统中,空时自适应处理(STAP)可有效抑制杂波并实现动目标检测。基于统计的STAP算法通过平稳的训练样本来估计检测单元内的杂波协方差矩阵,并设计相应的滤波器以提高检测单元的输出信杂比。但训练样本的平稳性在实际快变的杂波环境中无法保证,因而此类算法在实际非均匀杂波环境中性能较差。该文通过挖掘检测单元数据在角度-Doppler域上的稀疏性,提出一种新的直接数据域STAP算法。该算法通过稀疏恢复来获得检测单元的高分辨空时谱估计,有效地避免杂波旁瓣对目标检测的影响,进而实现不经过杂波抑制而直接运动目标检测的目的。同时由于不使用训练样本,可很好地避免训练样本内的非均匀性,该算法在实际非均匀杂波场景中有更广泛的应用前景。

Direct data domain STAP algorithm using sparse recovery

Space-time adaptive processing(STAP) is an effective tool for detecting moving targets in spaceborne/airborne radar systems.Statistical-based STAP methods need stationary training data to estimate the clutter covariance matrix and to design the corresponding adaptive filter to improve the output SCR(signal-clutter-ratio).However,the characteristics of the stationary environment are destroyed in the actual heterogeneous scenario,which causes performance degradation in statistical methods.This analysis exploits the sparsity of the received data in the angle-Doppler domain in a direct data domain STAP algorithm using sparse recovery to estimate the high-resolution space-time spectrum.In this way,this method effectively avoids the effect of the clutter sidelobe on the moving target so the moving target detection can be carried out directly on this high-resolution spectrum without clutter suppression.In addition,there is no heterogeneity introduced by the training data since only the test cell is needed,so this algorithm has great potential in actual heterogeneous clutter scenarios.