1Department of Computer Engineering, Ferdowsi University of
2Young Researchers and Elite Club, Mashhad Branch, Islamic Azad University, Mashhad, Iran
Clustering is an effective approach for organizing network nodes into hierarchical topology, aggregating sent data to the base station and prolonging the network lifetime. However, it may cause sudden death of nodes in some network regions, i.e., hot spots, due to heavy traffic load leading to disruption in network services. This problem is common for data collection scenarios in which Cluster Heads (CHs) have the responsibility of gathering and relaying information. To balance the energy consumption in network, the CH role must be rotated among all nodes and the cluster size should be determined such that uniformly distribute the workload over the CHs. In this paper, we propose a clustering algorithm that selects the nodes with highest remaining energy in each region as candidate CHs in order to pick the best nodes among them as final CHs. To consider the hot spot issue it employs fuzzy logic in order to adjust the cluster radius of CH nodes based on some local information (distance to base station and local density). Simulation results show that the proposed approach mitigates the hot spot problem and achieves an improvement in terms of network lifetime.