Adaptive pursuit learning for energy‐efficient target coverage in wireless sensor networks

Ramesh Upreti, Ashish Rauniyar, Jeevan Kunwar, Hårek Haugerud, Paal Engelstad, Anis Yazidi
2020 Concurrency and Computation  
With the proliferation of technologies such as wireless sensor networks (WSNs) and the Internet of things (IoT), we are moving towards the era of automation without any human intervention. Sensors are the principal components of the WSNs that bring the idea of IoT into reality. Over the last decade, WSNs are being used in many application fields such as target coverage, battlefield surveillance, home security, health care monitoring, and so on. However, the energy efficiency of the sensor nodes
more » ... in WSN remains a challenging issue due to the use of a small battery. Moreover, replacing the batteries of the sensor nodes deployed in a hostile environment frequently is not a feasible option. Therefore, intelligent scheduling of the sensor nodes for optimizing its energy-efficient operation and thereby extending the life-time of WSN has received a lot of research attention lately. In particular, this article investigates extending the lifetime of the WSN in the context of target coverage problems. To tackle this problem, we propose a scheduling technique for WSN based on a novel concept within the theory of learning automata (LA) called pursuit LA. Each sensor node in the WSN is equipped with an LA so that it can autonomously select its proper state, that is, either sleep or active, with an aim to cover all targets with the lowest energy cost possible. Our comprehensive experimental testing of the proposed algorithm not only verifies the efficiency of our algorithm, but it also demonstrates its ability to yield a near-optimal solution. The results are promising, given the low computational footprint of the algorithm. K E Y W O R D S adaptive pursuit learning, energy efficiency, learning automata, minimum active sensors set, target coverage, wireless sensor network INTRODUCTION With the explosive development of the Internet of things (IoT) technologies, various applications have surged up such as smart city, smart homes, smart hospitals, smart transportation, and so on. 1,2 We are witnessing the era of complete automation without any human intervention. All these applications of IoT are possible only because of the use of sensors. Sensors are the principal components that bring the idea of IoT into reality. 3 However, these sensors are an integral part of the wireless sensor networks (WSNs). The United States military first developed WSN technology as a sound surveillance system to detect and track submarines in the 1950s. 4 Since then WSN has gone through a vast amount of transitions and This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. thus it has become a practical solution for many applications such as target tracking, 5 smart factory, 6 surveillance, 7 target coverage, 8 industrial automation, 9 and so on. A WSN network comprises of the vast amount of sensors which brings together the sensed data of various activities of the physical world to give us meaningful information. 10 It should be noted that these sensor nodes in a WSN network operate autonomously with small batteries, which lasts from a number of months to years. Recharging or replacing the batteries of these sensor nodes is not a feasible option, especially if it is deployed in a hostile environment such as chemical reactors, underground tunnels, nuclear plants, and so on. 11 Therefore, the energy efficiency of these sensor nodes in WSN is a major concern and has attracted a great deal of attention from academia to industry. In this regard, the primary intention of this article is to expand the network life-time of the whole WSN system. One of the possible ways through which network life-time could be enhanced is by effectively covering the target in an energy-efficient way. The network life-time is defined in the context of network coverage problems as the duration of time elapsed from the network starts functioning with full coverage from its initialization to the time instant where the coveted coverage criteria is unsatisfied. 8 Target coverage is one of the important areas of the WSNs. It has been widely used for monitoring purposes, and it has its main application in military surveillance. 12 This coverage area can be defined as the area within which a sensor node can track the activities of the specified target. The mobile sensor deployment problem and the target coverage problem in mobile WSNs are NP-Hard. 13 The target coverage problem by the sensor nodes in the WSN includes three families of problems which are defined as follows according to Reference 14: • Area coverage: This coverage problem is concerned with the monitoring of the targets in the entire area of the network. • Target coverage: This coverage problem is concerned with the monitoring of only certain targets within the specified region of the network. • Barrier coverage: The barrier coverage problem aims at minimizing the probability of undetected penetration through the barrier in the network. The sensor node has a sensing area coverage based on its sensing range. Moreover, the sensor node also has a radio area coverage based on its communication range. 15 Intuitively, the specified monitoring field consists of densely deployed sensor nodes to avoid the formation of the coverage holes in the network. Particularly for the harsh terrains which are difficult to access, the sensor nodes are deployed in a random manner using an aircraft. When deployed randomly, usually, the areas covered by these sensor nodes overlap with each other. Due to the dense random deployment of the sensor nodes, there might be the sensing area that completely overlaps with the sensing area of other redundant sensor nodes. In WSNs, it is not advised to have multiple redundant sensors to cover the same target or area if it is already covered by other sensor nodes. Covering the same target or area by the redundant sensors affect the network-lifetime to a greater extent. Therefore, there is a multitude of research addressing the issue of identifying the redundant sensors and intelligently scheduling them in distinct time slots. Usually scheduling is done in such a way that a sensor can alternate between "Active" and "Sleep" mode to meet desired coverage requirements. 16
doi:10.1002/cpe.5975 fatcat:3hj273ok6rdjxblikzacqjxar4