Capon cepstrum weighted l2, 1 minimization for wideband DOA estimation with sonar arrays

In this paper, the wideband Capon cepstrum weighted l2, 1 minimization algorithm (WB-CW l2, 1) is presented for wideband direction of arrival (DOA) estimation using an acoustic array. The problem of wideband DOA estimation is converted into the estimation of the sparsity pattern of a jointly sparse signal and then solved by the weighted l2, 1 norm minimization. WB-CW l2, 1 uses the Capon cepstrum to calculate the support related weights. The basic idea of WB-CW l2, 1 is to assign large weights to those entries whose row indices are more likely outside the row support of the jointly sparse signals, so that their indices are more likely expelled from the row support of the solution. By doing so, WB-CW l2, 1 further enhances the sparsity of the solution and improves the DOA estimation accuracy, compared with l1-SVD, which is a typical algorithm based on unweighted l2, 1 norm minimization. Experimental results based on real sonar data show that the proposed algorithm outperforms l2, 1-SVD in terms of DOA estimation accuracy.