K杂波下基于Bootstrap的分布式检测

Bootstrap Based Distributed Detector Under Correlated K-Distributed Radar Clutter Background

传统雷达目标检测算法在相关K分布杂波下虚警概率过高,为此提出了一种基于Bootstrap的分布式检测(BBDD)算法。仿真数据表明,BBDD在相关K分布杂波中的虚警控制能力优于分布式广义符号检测器(DGSD),且K分布杂波形状参数变化对BBDD的虚警概率影响不大,适当增加BBDD脉冲参考单元数可更好地控制虚警概率。同时,为定量控制BBDD虚警概率,分块数据长度选择应根据杂波相关长度与脉冲累积单元数之间大小关系谨慎选择。最后,仿真结果证明了所述方法和相关结论的正确性。

To overcome the high probability of false alarm（PFA） of conventional radar detector under the correlated K-distributed clutter,a bootstrap based distributed detector（BBDD） was proposed in this paper.It has much higher capability to keep CFAR than the distributed generalized sign detector（DGSD） under the correlated K-distributed clutter.Besides,the PFA of BBDD is almost unchanged with the change of K-distribution shape parameter.The PFA of BBDD will decrease with the increase of the size of the reference sample number per pulse.The block size should be carefully chosen according to the correlated length of the clutter and the number of the cumulative pulse.Finally,simulation results are given to demonstrate the effectiveness of the proposed method.