摘要
MUSIC等为代表的高分辨DOA算法通常以精确已知的理想阵列模型为前提,由于实际天线阵列的位置误差必然存在,因此算法的性能必将受到严重的影响,甚至失效.本文通过深入分析阵列位置误差对阵列接收数据协方差矩阵的影响,提出了一种基于对接收数据协方差矩阵作Toep litz预处理来校正阵列位置误差的方法,同时结合特征值重构方法进行联合迭代运算,以便更加有效地抑制阵列位置误差的影响.理论分析和计算机仿真表明,所提方法能够有效地改善MUSIC算法的稳健性,提高多目标信号的角度分辨能力.
High resolution array direction finding techniques（such as the MUSIC algorithm） are usually used under the assumption that the array sensor locations are known precisely.However,the sensor location uncertainties always exist in practical circumstances.When the array sensor locations are randomly perturbed,the performance of the class algorithm then tends to deteriorate greatly and even becomes invalid.In this paper,a method based on the Toeplitz pre-processing of the covariance matrix is proposed.In order to enhance the capabilities further,an iterative algorithm is proposed to iteratively reconstruct both the Toeplitz and eigenstructure from the covariance matrix.Theoretical analysis and simulation indicate that the proposed algorithm improves the robustness of the MUSIC algorithm efficiently,as well as the multi-signal direction resolution.
出处
《西安电子科技大学学报》
EI
CAS
CSCD
北大核心
2011年第2期93-98,共6页
Journal of Xidian University
基金
国家自然科学基金资助项目(61072107)
西安邮电学院中青年科研基金资助项目(109-0412)
博士后基金资助项目(20090451252)
陕西省工业攻关资助项目(2009K08-31)
中央高校基本科研业务费专项基金资助项目(JY10000902025)
