DOA Estimation of Noncircular Signal Based on Sparse Representation

In this paper, we propose a novel method employing subspace fitting principle for DOA estimation of noncircular signal based on the sparse representation technology. The proposed method combines the signal information contained in both the covariance and elliptic covariance matrix of the received data matrix. We use the eigenvalue decomposition of the extended covariance to obtain the signal eigenvectors, and represent the steering vector on overcomplete basis subject to sparse constraint in subspace fitting method. After casting multiple dimensional optimization problem of the classical subspace fitting method as a sparse reconstruction problem, we use L1-norm penalty for sparsity, and optimization by the second order cone programming framework to obtain the DOA estimates. The proposed method can be used in arbitrary array configuration. Compared with the existing algorithms, the simulation results show that the proposed method has better performance in low SNR. Compared with L1-SVD, the proposed method also own better resolution probability.