IEEE Access (Jan 2024)
Harris Harks Optimization Based Clustering With Fuzzy Routing for Lifetime Enhancing in Wireless Sensor Networks
Abstract
In wireless sensor networks (WSNs), energy preservation through clustering and routing protocols has been verified as a significant effective method to extend the network lifetime. However, hot spot problem, excessive message overhead used for cluster formation, and frequent cluster maintenance are still the main challenges in clustering and routing protocols. The objective of this paper is to overcome these problems to enhance the network lifetime by proposing a new protocol combined Harris Harks Optimization Clustering with Fuzzy Routing named as HHOCFR. In HHOCFR, an improved Harris harks optimization (HHO) algorithm is utilized to select the best cluster heads (CHs) and form optimal clusters simultaneously by a novel encoding mechanism. Moreover, good point set based population initialization and neighborhood centroid opposition based learning mechanism are used to accelerate convergence and avoid being trapped into local optima. Thus, no extra message is needed for cluster formation. In addition, a fuzzy logic system (FLS) considers residual energy, energy consumption deviation of cluster, and distance to crossover point as descriptors to make decision on relay CH determination for each CH, resulting in energy balance among CHs so as to mitigate the hot spot problem. Thereafter, the CHs monitor the energy of the clusters to determine whether the rotation of CHs or re- clustering is started to maintain the clusters, which further decreases energy consumption of the network. According to the results, the average network lifetime of HHOCFR has increased by 34.20%, 28.11%, 23.65% and 17.43%, compared to IBRE-LEACH, DAPFL, IHHO-F and HHO-UCRA. For network throughput, HHOCFR is 12.7%, 13.51%, 20.95%, 10.87% higher than IBRE-LEACH, DAPFL, IHHO-F and HHO-UCRA. In addition, he energy consumption of HHOCFR is lower than IBRE-LEACH, DAPFL, IHHO-F and HHO-UCRA by 42.73%, 23.74%, 25.58%, 23.23%, respectively.
Keywords
