%0 Journal Article
%T 采用时延估计的外辐射源雷达杂波抑制算法<br>A Clutter Suppression Method Utilizing Time Delay Estimation for Passive Radars
%A 翟永惠
%A 吴江
%A 王鼎
%J 西安交通大学学报
%D 2015
%R 10.7652/xjtuxb201512008
%X 针对外辐射源雷达目标探测中监测通道存在分数倍时延杂波而造成杂波抑制性能下降的问题,提出了采用加权子空间拟合时延估计的外辐射源雷达杂波抑制算法(WSF？？TDE？？CM)。在假设接收数据中的目标信号远远弱于杂波信号的前提下,该算法首先利用接收数据的自相关矩阵通过加权子空间拟合的方法建立杂波时延估计模型,将分数时延估计问题转化为复正弦频率估计的优化问题,然后利用量子粒子群算法求解杂波时延,最后通过估计的杂波时延构造杂波矩阵,将接收信号投影到杂波空间的正交补子空间中,从而实现杂波的抑制。WSF？？TDE？？CM算法不需要设置滤波器阶数,在杂波时延为分数时延的情况下仍能保持良好的杂波抑制性能。仿真实验表明,当监测通道存在分数倍时延杂波时,WSF？？TDE？？CM算法与扩展相消算法相比,其杂波抑制比提高了约20 dB;同时,在目标回波信噪比为-30 dB时也能很好地检测到弱目标回波。<br>A novel clutter suppression method utilizing weighted subspace fitting time delay estimation (WSF？？TDE？？CM) is proposed to improve the performance degradation when the clutter has fractional time delay in external illuminators based radar. The method establishes a time delay estimation model by exploiting the weighted subspace fitting algorithm under the assumption that the target echo is far weaker than the clutter in the receiving signal. Then the model is solved by the quantum？？behaved particle swarm optimization to obtain the estimated time delay, which is utilized to construct the clutter matrix. Finally, the clutter suppression is implemented by projecting the received signal to the orthogonal subspace of the clutter space. There is no need to set the order of filter in WSF？？TDE？？CM and the method keeps good cancellation performance even if the clutter has fractional time delay. Simulation results show that compared with the extensive cancellation algorithm, WSF？？TDE？？CM improves the clutter attenuation by 20 dB when the clutter has fractional time delay, and yields quite good weak target detection performance even when the SNR equals －30 dB
%K 外辐射源雷达
%K 杂波抑制
%K 分数时延估计
%K 加权子空间拟合
%K 量子粒子群算法&lt
%K br&gt
%K external illuminators based radar
%K clutter cancellation
%K fractional time delay estimation
%K weighted subspace fitting
%K quantum？？behaved particle swarm optimization
%U http://zkxb.xjtu.edu.cn/oa/DArticle.aspx?type=view&id=201512008