基于Toeplitz降维子矩阵的空间多目标跟踪算法  

Tracking algorithm for multiple spatial targets based on Toeplitz submatrix of reducing dimension

作　　者：王安定[1] 裘渔洋[2] 王秀萍[1] 余燕平[1] 李式巨[3] 

机构地区：[1]浙江工商大学信电学院,浙江杭州310018 [2]浙江工商大学统计与数学学院,浙江杭州310018 [3]浙江大学信息与电子工程学系,浙江杭州310027

出　　处：《浙江大学学报（工学版）》2012年第4期739-743,769,共6页Journal of Zhejiang University：Engineering Science

基　　金：浙江省自然科学基金资助项目(Y1090232;Y6110639);浙江省教育厅基金资助项目(Y201017322)

摘　　要：为了克服传统的移动多目标跟踪计算量大、实时性差的缺点,针对阵列天线提出新的多目标跟踪算法.该算法利用阵元接收数据构造空域Toeplitz矩阵,截取该矩阵的一个子矩阵,采用Lagrange极值定理迭代跟踪子矩阵的最小噪声特征向量.根据得到的噪声特征向量,估计空间多个移动目标的波达方向(DOA).该算法对原接收矩阵进行了降维,因而有效地降低了计算复杂度.此外,对数据进行的Toeplitz重构,使得降维后的子矩阵保留了所有阵元的接收信息.仿真结果表明,该算法不但可以处理相关信源,而且具有很好的空间分辨率,适合对目标源进行实时跟踪.A new multiple target tracking algorithm for array antenna was proposed in order to overcome the drawbacks of heavy computational cost and poor real-time performance of the traditional moving multiple targets＇ tracking.The spatial Toeplitz matrix was constructed by the elements＇ receiving data in the algorithm and one of its sub-matrix was selected to iteratively track the extreme minimum noise eigenvector based on the Lagrange＇s theorem.The directions of arrival（DOAs） of spatial multiple targets were estimated according to the corresponding noise eigenvector.Due to the reduced dimension of the original matrix,the algorithm effectively lowered the computational complexity.Moreover,the Toeplitz reconstruction of the data kept the reduced matrix preserving the information of all elements.Simulation results show that the algorithm is not only an efficient way to deal with the coherent sources but also has the high spatial resolution,and it is suitable for tracking the spatial fast moving targets in real-time.

关 键 词：阵列天线 TOEPLITZ矩阵 特征空间 多目标跟踪 

分 类 号：TN911.3[电子电信—通信与信息系统]