A hybrid bio-inspired approach for clustering and routing in UWSNs using MPA and HGS

Abstract

Underwater Wireless Sensor Networks (UWSNs) encounter serious challenges due to dynamic topology, energy constraints, and high latency underwater communication. Existing methods for clustering and routing often fail to strike an optimal balance between data delivery reliability, energy efficiency, and latency reduction. This paper overcomes these shortcomings by developing a hybrid model that integrates the Hunger Games Search (HGS) and Marine Predator Algorithm (MPA) for improved clustering and routing in UWSNs. The MPA was chosen due to its stability in selecting the first sensors/cluster heads and creating the clusters, drawing inspiration from the foraging strategies of marine predators, which guides it extensively in the balance of exploration and exploitation. Simulations demonstrate that the proposed method achieves significantly better results than classical methods. In particular, the HGS-MPA framework consumes 26.6 % less energy than GWO-PSO, increasing network lifetime by 22.1 % (FINOD) and 15.8 % (HANOD). The packet delivery ratio is improved by 3.1 % against the following best-performing method, reaching 92.4 %. A statistical test performed with ANOVA showed that these improvements are statistically significant at P < 0.001. The results reinforce how the HGS-MPA framework would help improve energy efficiency, network lifetime, and communication reliability in UWSN systems.