AN ENHANCED K-MEDOID AND HYBRID OPTIMIZATION ALGORITHM FOR IMPROVING CLUSTER-BASED ROUTING PROTOCOL IN WIRELESS SENSOR NETWORKS

Abstract

Wireless Sensor Networks (WSNs) offer a robust technology for sensing and transmitting data across large geographic areas. However, restricted power, communication, and processing capabilities in WSNs can have a substantial influence on the network's longevity. To overcome these restrictions and enhance the energy efficiency of WSNs, sensor nodes are typically connected in groups, and the shortest path is established. Over the past years, many studies have been developed to cluster the nodes and choose the Cluster Head (CH) nodes. Among them, an Adaptive Sailfish Optimization (ASFO) with K-medoids outperforms in clustering and CH selection by reducing distance and latency among nodes. However, the limitation of K-medoids is that it may have difficulty establishing clusters due to the random selection of medoid locations. The random medoid initialization provides inferior results and degrades clustering quality. Also, the ASFO is presently facing the challenge of addressing optimization problems over extended periods. Hence, this article proposes a new hybrid with parallel processing energy-efficient cluster-based routing algorithm for WSNs with high energy efficiency and network longevity. The key goal of this study is to optimize the CH selection and achieve energy efficiency in WSNs. This algorithm comprises two major phases: CH selection and data transmission. First, the sensor nodes in WSN are clustered using the Fisher Median Naive Sharding (FMNS)-based K-Medoids scheme. Then, a new combination of Dove Optimization Algorithm (DOA) with SFO called DSFOA is utilized to select the optimal CHs. Furthermore, the tree-based multi-hop routing scheme is utilized to determine the shortest route for energy-efficient data transfer. The entire proposed work is named as Fisher K-medoids with Amalgamation of Dove and Sailfish Optimization-based Energy-Efficient Clustering and Routing Algorithm (FKADSO-EECRA). Finally, simulation results exhibit that the FKADSO-EECRA achieves high Packet Delivery Ratio (PDR), throughput, and network lifetime while reducing End-to-End Delay (E2D), jitter, energy consumption, and routing overhead compared to the conventional cluster-based routing algorithms.