基于MTI滤波器特性的改进直接数据域算法的研究

Study on modified direct data domain STAP based on MTI filter characteristic

传统的空时自适应处理方法属于统计方法,一般假设其它相邻距离单元的训练样本满足独立同分布的条件,利用这些训练样本来估计待检测单元中杂波的协方差矩阵.对机载雷达来说,严重非均匀环境很难满足样本独立同分布的要求.直接数据域方法只利用待检测距离单元本身的数据来获取训练样本,因此得到广泛应用.提出一种改进直接数据域算法,在利用空时窗滑动对消目标信号时,基于滤波原理,对空时窗的权系数进行优化,这样有利于在对消目标的同时保留更多杂波信息,进而求解的自适应权值对杂波的抑制性能更佳.仿真结果表明了改进算法的有效性,相比原始直接数据域算法,具有更窄的改善因子凹口,提高了对慢速目标的检测性能,且计算量没有增加.

Among most of the conventional space-time adaptive processing （STAP） approaches, statistical methods are used, by which data from adjoining range cells must be used to estimate the covariance matrix of the clutter in the cells to be examined and also it is presumed that these data samples are independent and identically distributed. However, this requirement is difficult to satisfy because of the highly inhomogeneous environments in the airborne radar application. For this reason, the direct data domain （DDD） STAP is widely applied, with only the primary data needed. In this paper, a modified version of DDD STAP is presented, which makes use of the sliding of space-time windows to cancel the target signals. Based on the filtering principle, the weighting coefficient of the space-time windows can be optimized,which is not only beneficial to target cancelling but can also preserve more information from clutter, making the performance of weak target detection become better. Simulated numerical experiments demonstrate the novel method is very effective and has a better performance than the original DDD STAP, and its computation load doesn＇t increase.