A weighted Markov-clustering routing protocol for optimizing energy use in wireless sensor networks

Abstract

Interest has increased in wireless sensor networks (WSNs) in fields such as healthcare, industrial control and environmental monitoring. More recently, WSNs have been widely deployed in Internet of Things (IoT) based applications. They are considered as an essential part of IoT networks. The expanded use of WSNs raises daunting technical challenges, not the least of which is sensor battery energy conservation. The design of efficient sensor clustering strategies to reduce energy consumption by data transmission throughout the WSN has become crucial. Several applications of WSNs induce a random deployment of sensors such as the monitoring of conflict zones, the study of natural phenomena in hostile zones or rescue operations. In this context, energy conservation should be ensured according to the non-uniformity of the resulting sensor distribution in different areas of the network. In this study, a novel weighted Markov clustering protocol that considers sensor abundance for cluster-head and queried sensor selections is presented. The new protocol aims to decrease intra-cluster energy consumption by reducing the sending of redundant data in sensor-dense regions. In addition, it attempts to prolong sensor-sparse regions lifetime by limiting the number of queried sensors. This protocol combines a Markov clustering of sensors with a sensor weighting based on residual energy and sensor abundance in the network. The proposed protocol is a significant improvement of an existing unweighted Markov clustering protocol. The unweighted Markov clustering protocol is based on sensor residual energy and sensor location without taking into account the sensor abundance in different areas of the network. Simulations affirm that the new protocol handles more appropriately the non-uniformity of sensor distribution and enhance the durability of wireless sensor networks. Indeed, simulation results show that the proposed clustering protocol outperforms its unweighted ancestor and other well-known clustering protocols in terms of energy conservation and network lifetime. The number of expired sensors and the average dissipated energy are reduced, whereas, the average sensor lifetime is prolonged compared to the unweighted ancestor, or HEED, LEACH, PEGASIS, and TEEN.

Keywords

• WSN
• Markov clustering (MCL)
• Sensor abundance
• Sensor distribution
• Routing protocol
• Energy consumption