An Improved Nature Inspired Algorithm to Design Energy Efficient and Stability Aware Routing Protocol for Wireless Sensor Networks

Abstract

Nature-inspired algorithms represent a category of highly potent optimization techniques that draw inspiration from natural processes. They have found applications across a broad spectrum of research disciplines. These algorithms are characterized by their simplicity, adaptability, and flexibility, making them suitable for various problem-solving scenarios. These algorithms have existed for generations but have garnered significant interest from researchers over the past two decades. In this thesis, several nature-inspired algorithms have been studied, including CS, GWO, NMRA, SMA, WOA, COOT optimization, and others. These algorithms are recognized as state-of-the-art and have demonstrated their competitiveness and applicability across various research domains. In addition to their advantages, these algorithms exhibit certain drawbacks, including algorithm trapping in local optimal solutions, limited exploration capabilities, slow convergence rates, and complex parameter tuning. To overcome these limitations, it is imperative to design hybrid or enhanced versions of algorithms capable of effectively addressing high-end optimization problems. It has become evident that no single algorithm is universally suited for all optimization challenges. Therefore, the central objective of this work is to introduce an algorithm capable of successfully solving standard IEEE CEC benchmark problems and subjected to various dimensional and statistical analysis to validate results. In this work, a number of hybrid algorithms have been developed inspired from the exploitation capability of NMRA with added features of self adaptability and stagnation avoidance phase.

The widespread use of wireless sensor devices and their advancements in terms of size, deployment cost, measurement of environmental events and user friendly interface have given rise to many applications of WSNs. Such networks need to utilize routing protocols to forward data samples from event regions to sink. The focus of such routing protocols is to forward collected data samples via minimum cost links (in terms of energy consumption and time). Clustering is an efficient data aggregation method that effectively reduces the energy consumption by organizing nodes into groups (clusters). The proposed algorithms have been applied to create cluster-based routing protocols for WSNs. Most previous work on WSNs has focused on minimizing energy consumption during data extraction processes. This emphasis is because SNs operate on batteries that deplete rapidly with each operation. Therefore, proper CH selection and their load balancing using efficient routing protocol is a critical aspect for the long run operation of WSN. In light of this, the research developed nature-inspired algorithm-based clustered routing protocols and minimized the network's energy resource consumption, extending its overall lifetime with consideration for static and mobile SNs.