Distributed Compressed Sensing Algorithm for Cluster Architectures of WSNs

Abstract

According to the traditional CS theory, each sensor node in the wireless sensor networks always is assumed to directly deliver relative information to sink node, which only considers the intra-signal correlation structure. In addition, these may lead to the loss of the node information and the overenergy consumption. To adjust the processing power and the energy limitation of the node, combined with the inter-signal correlation structure and the joint sparsity models - JSM1, this paper presents a new distributed compressed sensing algorithm for cluster architectures of wireless sensor networks; the proposed algorithm reconstructs the nodes based on side information. Simulation analysis shows that the improved distributed compressed sensing algorithm not only can access to the accurate reconstruction of the nodes, but also can reduce energy consumption during the process of algorithm greatly.