Sample selection strategy in DFT based compressive sensing

A recently proposed strategy of selecting samples for a unique reconstruction of a signal is analyzed in this paper. The considered signal is sparse in the discrete Fourier transform (DFT) domain. Since the problem is of theoretical importance, we use the basic direct search method for the reconstruction and comparisons of sampling strategies. It is shown that, by using the proposed sampling strategy method, the sparsity limit for the unique reconstruction is increased in comparison to the random selection of samples.