A Green Distributed Signal Reconstruction Algorithm in Wireless Sensor Networks

Abstract

Environmental consideration provides new trends in wireless communication network known as green communication. The main object of green communication is to save as much as possible the energy consumption of the communication system. In this paper, the authors have investigated the green distributed nonlinear state estimation problem in wireless sensor networks (WSNs), which will be seamlessly integrated with the forthcoming 5G communication system. A distributed signal reconstruction algorithm is proposed by employing compressive sensing and consensus filter to solve sparse signal reconstruction issue in WSNs with energy efficiency considered. In particular, the pseudo-measurement (PM) technology is introduced into the cubature Kalman filter (CKF), and a sparsity constraint is imposed on the nonlinear estimation by CKF. In order to develop a distributed reconstruction algorithm to fuse the random linear measurements from the nodes in WSNs, the PM embedded CKF is formulated into the information form, and then the derived information filter is combined with the consensus filter, while the square-root version is further developed to improve the performance and strengthen power saving capability. The simulation results demonstrate that the sparse signal can be reconstructed with much fewer nodes in decentralized manner and all the nodes can reach a consensus, while providing some attractive benefits to the green 5G communication system.

Keywords

• Green
• compressive sensing
• nonlinear
• cubature Kalman filter
• square-root decomposition
• consensus filter