微气候环境模拟试验室用温度控制系统设计-西南大学.DOC

DOI: 10.13718/ki.xdzk.2015.3.24 (第x篇) 基于Hankel矩阵分解的互素阵列高分辨目标定向谭伟杰，冯西安，张杨梅摘 要：互素阵列经过Khatri-Rao（KR）积处理可以有效扩展阵列虚拟孔径，的阵列模型等效为一个确定噪声下的单快拍的相干源估计问题。该文提出一种基于 Hankel 矩阵分解的方法，采用来进行高分辨定向。首先将阵列采样协方差数据进行Khatri-Rao积预处理，进行相关数学变换后，用去冗余后的数据构建 Hankel矩阵，然后采用奇异值分解，转换为多项式求根，最后找出感兴趣目标位置，并通过最小二乘法来求解出目标幅度。该方法不仅解决了处理后转换为的相干源问题，传统MUSIC方法中的网格失配问题，并且对于近目标具有良好的分辨能力。仿真结果表明该方法的有效性和优越性。关键词： Hankel矩阵分解；互素阵列；网格失配；高分辨目标定向；近空间目标 中图分类号：TN911.7　　文献标志码：A 文章编号：1673-9868(2015)3-0001-07 High Resolution Target Direction Finding Based on Hankel Matrix Decomposition for Coprime Array Tan ei-jie　Feng i-an Zhang Yang-mei School of Marine Science and Technology, Northwestern Polytechnical University, Xi’an 710072, Shaanxi, China Abstract: Co-prime array has large virtual array aperture by Khatri-Rao product processing，but the array model after processing is equivalent to coherent sources estimation under determined noise on a single snapshot. In this paper, a high resolution direction finding method based on Hankel matrices decomposition and Polynomial rooting is proposed. Firstly, the covariance of the array data is preprocessed by KR product, repeated rows are removed. The obtained data is used to build a Hankel matrix. Secondly, perform singular value decomposition on the Hankel matrix, the problem is transformed into the determination of polynomial roots to find the interested target location. Finally, the amplitude of the target echo is solved by using the least squares method. To some extent, the proposed method can overcome grid mismatch which would be a problem in the traditional MUSIC method, compare with smoothing technology, reduce the computational complexity, and has good resolution even for closely spaced target. Computer simulations verify the effectiveness and superiority of the new algorithm. Key words: Hankel matrix decomposition; co-prime array; grid mismatch; high resolution target direction finding; closely spaced target 1引言 [1]。在阵列信号处理中通过增加阵列天线Vaidyanathan 和Pal等提出的具有稀疏结构的互素阵[2][3]