ESAC: An Energetic Sustainable Adaptive Clustering Protocol for Heterogenous Wireless Ad Hoc Networks (HANETs)

Abstract

We study adaptive clustering in heterogenous wireless ad hoc networks (HANETs) and propose ESAC, an energetic sustainable protocol that adaptively forms clusters using a weighted election function based on the current energy status of nodes in each region. The proposed protocol is distributed, and each node independently makes the decision to become a cluster head at the current round. The workload of a cluster head is heavy, including forming the cluster, collecting data from cluster members, data aggregation and transmission to faraway base station. ESAC selects an optimum number of cluster heads in each round among high-powered nodes that can afford the heavy duties of a cluster head. The protocol keeps track of the nodes' residual energies and selects the high-powered nodes as cluster heads more frequently than other nodes. As a result, although each node starts with a different initial energy, it adaptively tunes its energy consumption with its residual energy and the current energy status of other nodes, so that all nodes deplete their batteries at almost the same time. ESAC is energetic sustainable and the nodes last for as long as possible. Classical clustering protocols assume the same or certain number of initial energy levels for nodes, however ESAC is designed for HANETs where nodes have different energy levels. We validated the effectiveness and efficiency of our protocol through simulations. The analysis of our results showed that on an average ESAC reduces the instability into 46% and improves the total throughput by 42% and the stable phase throughput by 21% (and as high as 32%) in a HANET.