Distributed parameter estimation for mobile wireless sensor network based on cloud computing in battlefield surveillance system

Abstract

The construction of a battlefield surveillance system is very important for monitoring the attack of enemy aircrafts and missiles, which integrates various sensors and mobile devices. Then, multiple battlefield surveillance systems can be connected together to form a battlefield surveillance network. The mobile nodes can be deployed in a certain region to monitor enemy aircrafts and missiles. Thus, some important issues have to be solved efficiently, including the cooperation across the administrative domains of a cloud network, the direction-of-arrival (DOA), and a polarization estimation algorithm for a mobile wireless sensor network (MWSN). In this paper, the architecture of a battlefield surveillance system is constructed based on mobile cloud computing and 5G link. The root multiple signal classification (Root-MUSIC)-like algorithm is proposed for estimating the 1-D DOA and a polarization parameter with a uniform linear array. The Root-MUSIC algorithm is replaced by the Fourier transform, the former algorithm that can be extended to an arbitrary topology structure of a MWSN. Then, the proposed algorithm is extended to the 2-D DOA and a polarization estimation in further. Based on the deployment of different MWSNs, the estimation results of DOA and polarization parameters are fused in order to improve the estimation performance. Finally, the parameter information (DOA and polarization parameter) of enemy aircrafts and missiles can be achieved. The computer simulation verifies the effectiveness of the proposed algorithm. The proposed algorithm ensures the parameter estimation accuracy with a low computational complexity.