GSA-RPI: GSA based Rendezvous Point Identification in a two-level cluster based LR-WPAN for uncovering the optimal trajectory of Mobile Data Collection Agent

Abstract

Energy efficient data collection with minimum delay is very essential for IoT applications using sensors attached to Low Rate Wireless Personal Area Networks (LR-WPAN). The nature and position of the data collection and processing point, termed as Sink, plays a vital role in the data reliability, delay, and network lifetime. The static sink causes hot-spot problem to occur in Wireless Sensor Network (WSN), since the nodes nearer to the sink may die quickly. The Mobile Data Collection Agent (MDCA), commonly known as Mobile Sink (MS), can be used to mitigate this problem. A two level clustering scheme with an optimal data collection strategy is proposed in this work. In the first level, hierarchical clusters are formed and the Cluster Heads (CHs) are selected based on the residual energy and connectivity of the devices. The CHs are partitioned into disjoint sets and each set is represented by a Rendezvous Point (RP) that covers every element in the set. In the second level clustering, a deterministic approach is used for the RP identification followed by a meta-heuristic approach known as Gravitational Search Algorithm (GSA) for RP optimization. Finally, the MDCA collects data by visiting RPs along the path obtained using Simulated Annealing (SA). The proposed method is compared with Particle Swarm Optimization based path finding, SPT-GDA, and BR-CTR algorithms. The algorithms are analyzed for number of CHs, RPs, path length, Average Residual Energy, and Network Lifetime in LR-WPAN.