JoWUA

Abstract

Attaining a prolonged network lifetime, maximized coverage, and high performance are vital design factors that have to be maintained in a Wireless Sensor Networks (WSN). Such factors are dependent on the stability and optimality of the protocol employed to formulate Sensor Nodes (SNs) into mutual clusters that effectively work around fulfilling specific performance goals. SEP is a heterogeneity-aware protocol implemented based on Low-Energy Adaptive Clustering Hierarchy (LEACH) protocol, designed to prolong the stability period of the network defined by the time interval before the death of the first SN. Nevertheless, adaptability in design is a scheme that has not been extensively applied in the formation of WSN election protocols. Adapt-𝑃 is a probabilistic model for adaptivity designed to evolve the probability of selecting a cluster-head based on the active status of the WSN, represented by the residual energy of and distances between SNs. However, the adaptive probability 𝑃adp formalized in Adapt-𝑃 is developed based on the remaining number of SNs 𝜁 and optimal clustering κmax, yet 𝑃adp does not implement the probabilistic ratios of energy and distance factors in the network. Furthermore, Adapt-𝑃 does not localize cluster-heads in the first round properly because of its reliance on distance computations defined in LEACH, that might result in uneven distribution of cluster-heads in the WSN area and hence might at some rounds yield inefficient consumption of energy. This paper utilizes 𝑘-means++ and Adapt-𝑃 to propose 𝑃c𝜅max-means++ clustering algorithm that better manages the distribution of cluster-heads and produces an enhanced performance. The algorithm employs an optimized cluster-head election probability 𝑃c developed based on energy-based 𝑃𝜂(𝑗,𝑖) and distance-based 𝑃𝜓(𝑗,𝑖) quality probabilities along with the adaptive probability 𝑃adp, utilizing the energy 𝜀 and distance optimality 𝑑opt factors. Furthermore, the algorithm utilizes the optimal clustering 𝜅max derived in Adapt-𝑃 to perform adaptive clustering through 𝜅max-means++. The proposed 𝑃c𝜅max-means++ is compared with the energy-based algorithm 𝑃𝜂𝜀𝜅max-means++ and distance-based 𝑃𝜓𝑑opt𝜅max-means++ algorithm, and has shown an optimized performance in term of residual energy and stability period of the network.