Sparsity Estimation Method in Compressed Data Gathering of Wireless Sensor Networks

In traditional compressed data gathering (CDG) methods of wireless sensor networks (WSNs), It is assumed that the sparsity of sparse signal is priori and stationary. This assumption is quite different from the practical application. WSNs are usually arranged in complex environments, and the sparsity is constantly changing with the change of environment. Hence, it is necessary to estimate sparsity in advance. In this paper, a relative threshold based sparsity estimation method (RTSE) is proposed to solve the problem of sparsity estimation from the measured signal. Compared with the traditional method, our method could improve the correct rate (CR) of sparsity estimation by adjusting the threshold and get the sparsity directly before the reconstruction of the whole sparse signal. Simulation results show that the proposed method can estimate signal sparsity with higher CR, compared with previous method.