Circular array direction-of-arrival estimation in modal space using sparsity constraint

Acoustical science and technology

Hai-Yan Song¹  Ming Diao¹  Chang-Yi Yang²  Jin-Ping Qin³ 

[1] College of Information and Communication Engineering, Harbin Engineering University;Department of Computer Science and Information Engineering, National Penghu University of Science and Technology;School of Electrical and Information Engineering, Heilongjiang Institute of Technology

关键词: Direction of arrival;    Compressive sensing;    Uniform circular array;    Phase mode;    Sparsity constraint;   

DOI  :  10.1250/ast.39.343

学科分类：声学和超声波

来源: Acoustical Society of Japan

【 摘 要 】

A uniform circular array (UCA) can provide 360° azimuthal coverage and can be steered in any direction in two-dimensional (2-D) space without changing the shape of the pattern. These attractive features led to the rapid development of direction-of-arrival (DOA) estimation techniques using a UCA. However, most of the previous work on DOA estimation based on a UCA only utilized the time-space statistical information available from the array signals and did not exploit the inherent sparsity of the underlying signal in space domains. In this paper, we develop a new circular array DOA estimation approach that can achieve spatial sparsity, and thus improve spatial resolution, by imposing penalties based on the l1-norm. Our approach differs from most other circular array DOA estimation methods not only in its recognition of the concept of phase modes but also in how it imposes the spatial sparsity constraint on the impinging signal wave fields to achieve better DOA estimation performance. It turns out that our approach, in essence, applies the compressive sensing technique to the modal array signal processing and does not need to reconstruct the correlation matrix or its inverse, so can work well when the sources are coherent. Computer simulations with several frequently encountered scenarios, such as a single source and two closely spaced coherent sources, indicate the superior DOA estimation resolution of our proposed approach as compared with existing techniques. In addition, from a statistical viewpoint, the performance of our proposed approach is investigated more closely by considering the root-mean-square error (RMSE) versus the signal-to-noise ratio (SNR), number of snapshots, or number of sensors, and its excellent DOA estimation accuracy is demonstrated.

【 授权许可】

Unknown