Abstract
In this article, the direction of arrival estimation (DOA) of strictly noncircular sources with unknown mutual coupling is considered for wireless sensor array network (WSAN), and then a joint weighted block sparse recovery algorithm based on weighted subspace fitting (WSF) is proposed for DOA estimation.. In the proposed method, two block sparse representation models associated with the WSF principle are firstly constructed to remove the influence of unknown mutual coupling. Then, combining the advantages of noncircularity, a joint weighted block sparse recovery scheme is proposed to estimate DOA, in which the noncircular MUSIC-like (NC MUSIC-Like) spectrum function is utilized to form a weighted matrix for enhancing the solutions sparsity. Finally, the desired DOAs can be achieved with the help of the reconstructed block sparse matrix. Extensive experiments are simulated to verify that the proposed method can achieve superior estimation performance under the condition of unknown mutual coupling.
Highlights
Wireless sensor array network (WSAN) technique is widely applied in the field of communications and radar during the last decades [1]
They are all rely on the eigenvalue decomposition (EVD) of covariance matrix, which indicates that those subspace-based methods may be unable to work properly in the challenging environment, such as limited snapshots or/and low signal-to-noise ratio (SNR) [6], [7]
For exploiting the noncircularity of the signals, a joint weighted subspace fitting (WSF) framework based on l0-norm penalty via combining the principle of block sparse recovery can be constructed as min ||T ||0 s.t
Summary
Wireless sensor array network (WSAN) technique is widely applied in the field of communications and radar during the last decades [1]. L. Li et al.: DOA Estimation of Strictly Noncircular Sources in WSAN via Block Sparse Representation. By applying a parameterized operation to the actual disturbed array manifold, a novel block-structure steering matrix is deduced to avoid the unknown mutual coupling. From the sparse recovery aspect, the particular selection matrix is put forward to eliminate the interference of unknown mutual coupling [24], [25] They all suffer from array aperture loss. In this article, considering both noncircular sources and unknown mutual coupling in WSAN, a weighted subspace fitting (WSF) framework via block sparse representation is presented to estimate DOA. Based on the principle of WSF, two block-structured sparse data models are structured to deal with the influence of unknown mutual coupling without losing the array aperture. M stands for an M × M dimensional exchange matrix with ones on the anti-diagonal and zeros elsewhere. || · ||0, || · ||1, || · ||2 and || · ||F represent l0-norm, l1-norm, l2-norm and Frobenius norm, respectively. det| · |, E[·] and tr{·} mean determinant operations, mathematical expectation and trace operation, respectively
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call