LAR-CH: A Cluster-Head Rotation Approach for Sensor Networks

Abstract

In wireless sensor networks, sensors are expected to work autonomously for a long time with a finite source of energy. Therefore, the sensors must be energy efficient in their duties to prolong the network lifetime. Cluster-based protocols are widely used in the literature to prolong the network lifetime. Clustering protocols use cluster-head rotation (CH-rotation) and re-clustering methods to rotate the energy-intensive load of CH nodes among other nodes of the network. Re-clustering is a global method, so its overhead cost is higher compared with CH-rotation. Existing clustering protocols employ an energy threshold on the residual energy of the CH for CH-rotation. This energy threshold can be static or dynamic. The static threshold does not consider the current energy load of the CH for CH-rotation. Thus, it increases premature death of CH nodes. In this paper, a load-aware rotation of CH (LAR-CH) approach is proposed, which sets a dynamic threshold for CH-rotation to reduce the premature death of CH nodes. The LAR-CH uses the current energy load of the CH to determine the dynamic threshold for CH-rotation. The simulation results show that LAR-CH reduces the premature death of CH nodes by 40% compared with the low-energy adaptive clustering hierarchy protocol.