A Robust Sparsity Estimation Method in Compressed Sensing

Compressed sensing has been widely used in wireless sensor networks. In compressed sensing field, many aspects depend on the sparsity of the sparse signal, and we usually assume that the sparsity is known in advance, but the sparsity is unknown and not fixed in practice. So it is necessary to estimate the sparsity before we use it. In this paper, we propose a new method to estimate the sparsity, we use greedy algorithm and relative threshold to estimate the sparsity. Comparing with the traditional method, our method does not need reconstruct the whole signal, needes fewer number of measurements and estimation times, has better performance in low SNR scenarios or when the signal is changing. The simulation indicate the advantages of the new method.