A sparsity feedback-based data gathering algorithm for Wireless Sensor Networks

• Published in: Computer networks (Amsterdam, Netherlands : 1999) Vol. 141; pp. 145 - 156
• Main Authors: Lv, Cuicui, Wang, Qiang, Yan, Wenjie, Li, Jia
• Format: Journal Article
• Language: English
• Published: Amsterdam Elsevier B.V 04.08.2018; Elsevier Sequoia S.A
• Subjects: Agents (artificial intelligence); Algorithms; Compressive data gathering; Compressive Sensing; Data recovery; Energy balance; Feedback; Graph theory; Monte Carlo simulation; Remote sensors; Sensors; Shortest-path problems; Sparsity; Sparsity feedback; Wireless networks; Wireless Sensor Networks; Energy balance; Compressive Sensing; Sparsity feedback; Wireless Sensor Networks; Compressive data gathering
• ISSN: 1389-1286; 1872-7069
• DOI: 10.1016/j.comnet.2018.05.022

As a means of detecting abnormal events in Wireless Sensor Networks (WSNs), this paper presents a Compressive Sensing (CS)-based algorithm, called Minimum Spanning Tree and Mobile Agent-based Greedy Shortest Path (MST-MA-GSP). The algorithm first of all uses a sparsity feedback mechanism to accurately estimate the sparsity k of the sensor measurements. It then uses Monte Carlo experiments to determine the minimum number of required measurements Mmin. According to the value of Mmin, the algorithm adaptively adjusts the number of measurements M in order to maximize its recovery performance. The experiments show that the proposed algorithm is superior to other compressive data gathering (CDG) algorithms in terms of energy balance, whilst the adaptive Mmin mechanism guarantees a reconstruction accuracy of at least 99%. Additionally, the sparse binary matrix used in the MST-MA-GSP algorithm offers better recovery of sparse zero-one data than other CDG-based measurement matrices.