Parameter estimation of coherently distributed sources using sparse representation

Abstract

In this paper, a new estimator of coherently distributed source employing the sparse representation technology is proposed by utilizing subspace fitting principle. The proposed method uses the eigenvalue-decomposition method on the sample covariance matrix of the sensor array received data and obtains the signal eigenvectors. We represent the generalized steering vectors of coherently distributed source containing central direction-of-arrival (DOA) and angular spread on over complete dictionaries subject to sparse constraint in subspace fitting method. Then subspace fitting problem is transformed into a sparse reconstruction problem. Finally, we use L1 norm method to solve the sparse reconstruction problem, which is optimized by the second order cone programming (SOCP) framework. Compared with the existing algorithms for coherently distributed source, such as DSPE and ESPRIT, the simulation results show that the proposed method has better resolution performance, especially in small number of snapshots.