Naked Mole-Rat Algorithm with Improved Exploration and Exploitation Capabilities to Determine 2D and 3D Coordinates of Sensor Nodes in WSNs

Abstract

Wireless sensor network (WSN) consists of sensor nodes deployed with limited battery power to observe and report certain events taking place in the sensor field. Location determination is a critical task, but an important parameter in various applications of WSNs. It is an important task to locate all the sensor nodes efficiently to get the accurate and precise location of an occurring event. Numerous applications require the actual determination of their exact location that includes surveillance, animal tracking, tracking of moving objects, military applications, and forest fire detection. There are various localization algorithms available in the literature to deal with the localization problem in WSNs. A single anchor node and six virtual anchors are placed and assumed to locate all unknown nodes deployed in 2D and 3D environments using various meta-heuristic algorithms available in the literature. In this paper, three versions of the Naked Mole-Rat Algorithm (NMRA) are proposed to enhance its exploration and exploitation capabilities. To check the effectiveness of NMRA variants, CEC 2019 benchmark functions are used for comparison with basic NMRA. The results show that proposed NMRA variants converge fast and provide an optimum solution for most of the benchmark functions. Also, the proposed NMRA variant outperforms well in terms of localization error in comparison to existing localization solutions for both 2D and 3D environments.