Density-based spatial clustering technique for wireless sensor networks

Abstract

Clustering in wireless sensor networks (WSNs) is an important stage for the communication between sensor nodes. Many clustering techniques were introduced in the literature with different characteristics. The main goal of them is to facilitate a power-aware communication between a large number of deployed nodes. Clustering techniques can be classified into two main groups; centralized and distributed techniques. Centralized techniques consume too much overhead, especially with a large number of nodes. Distributed techniques depend on probabilistic mechanisms. The main drawback of these techniques is that it does not guarantee a uniform distribution of clusters because it does not consider the topological characteristic of the deployed network. As a result, the total energy consumption increases and the network lifetime decreases. This work proposes a distributed density-based clustering technique called Spatial Density-based Clustering (SDC). It overcomes the drawbacks of other related distributed techniques. It aims to achieve balanced energy consumption all over the constructed clusters. Compared to other distributed clustering techniques, simulation results show that the proposed technique achieves less energy consumption and longer network lifetime.