Efficient sampling for radar sensor networks

Abstract

Compressive sensing (CS) is an excellent technique for data acquisition and reconstruction in radar sensor networks (RSNs) with a high computational capability. This paper presents a new efficient and effective signal compression and reconstruction algorithm based on CS principles for applications in real–world RSNs, in which the signals are obtained in real–world experiment of RSNs. The proposed algorithm neither requires any new optimisation method, nor needs complex pre–processing before compression. This method considers correlation between radar sensor signals to reduce the number of samples required for reconstruction of the original radar signals. We compare our algorithm's performance and complexity with some existing work, such as joint PCA & CS, DCS, and traditional CS. Numerical results show that the proposed algorithm performs more efficiently and effectively without introducing any more computation complexity. With more sensor nodes, our algorithm is more efficient, which significantly reduces the number of samples required per sensor.