IDA: Improved dragonfly algorithm for load balanced cluster heads selection in wireless sensor networks

Abstract

Efficient Energy Consumption and Network Lifetime are significant concerns in wireless sensor networks and allied disciplines. Clustering is one of the available solutions, but optimized cluster head selection is a prime issue nowadays. Many solutions have been given for solving this issue considering some attributes, but with time, meta-heuristics algorithms have become widely used for real-world applications. The nature-inspired algorithm is seeking researchers' wide attention as it gives the capability to self-learn and perform better. In this paper, we have proposed the Dragon Fly algorithm with multi-attribute decision-making inspired by the dynamic and static behavior of the dragonfly. The Proposed IDA (Innovative Dragonfly Algorithm) is a hybrid approach in which dragonfly and multi-attributes are combined for optimal cluster head selection. The proposed method is a way to compute multi-attributes of sensor nodes for ranking them and selecting optimized cluster heads. The energy consumption of IDA is 0.4896, NBA (Novel Bio-Inspired Algorithm) is 0.4321, FLPSOC (Fuzzy Logic and PSO-based energy efficient clustering) is 0.4421, and ESO-LEACH (PSO-based energy efficiency) is 0.4678 at which means proposed IDA is better than other compared algorithms in energy consumption. The throughput of IDA is 64.99, which is better than existing different compared algorithms. The number of alive nodes in the proposed method is, and that of compared algorithms is; thus, IDA has enhanced network lifetime compared to others. The IDA algorithm is compared with NBA, FLPSOC, and ESO-LEACH, validating that the proposed algorithm performs better than the classical and compared algorithm.