Markov Modulated Bernoulli Prediction Process‐based Cluster Head Selection Mechanism for improving resilience in wireless sensor networks

Abstract

Summary The clustering process in wireless sensor networks is considered to be predominant for facilitating minimized energy consumptions through the selection of potential cluster heads that are efficient in energy and trust. The significant cluster head selection is deterministic in topology control with the view to manage the sensor member nodes of the cluster with the view to enhance energy efficiency in the network. In this paper, a Markov Modulated Bernoulli Prediction Process‐based Cluster Head Selection Mechanism (MMBPP‐CHSM) is proposed by modeling the success probability of sensor nodes' state into a Bernoulli process that dynamically evolves over time based on Markov chain for extending the lifetime of the network. This proposed MMBPP‐CHSM scheme is vital in quantifying the reliability and availability of the sensor nodes in a dynamically evolving wireless sensor network through the estimation of trust and energy in order to facilitate effective cluster head selection. The simulation results of the proposed MMBPP‐CHSM scheme confirmed a superior enhancement in percentage of alive nodes and death nodes in the network to a maximum level of 28% and 23% on par with the baseline cluster head selection approaches used for investigation.