Cluster Size Optimization in Gaussian Distributed Wireless Sensor Networks
Vinay Kumar, Sanjay B. Dhok, Rajeev K. Tripathi, Sudarshan Tiwari · 2014
To deal with sensor network limitations such as limited energy and short range communication, sensor nodes are grouped into mostly non overlapping subsets called clusters. Choosing optimal number of clusters provides benefits such that limited resources can be utilized more efficiently and network lifetime is improved. Many of the existing researches provided the cluster size optimization in Wireless Sensor Networks (WSNs), in which nodes are uniformly and randomly placed in the sensing field (e.g. controllable WSN). Deployment of sensor nodes affects the energy consumption of WSNs along with individual nodes because the distance between nodes and Base Station (BS) is different due to different node position; consequently nodes have different energy loss. The energy efficient way of sensor deployment in sensing field is controlled node deployment with uniform distribution. However this procedure for node deployment may not be practically possible for some applications like, in large WSNs, locations of the sensing field may not be physically accessible because of geographical constraints. In this paper, we provide an analytical framework for the cluster size optimization of WSNs that follow Gaussian node deployment. This type of node deployment reduces energy hole problem, provides enhanced intrusion detection capability and support realistic applications. We have provided expression for optimal number of clusters using circular sensing model of nodes for square sensing field with consideration of boundary effect. We have also compared the cluster size optimization for uniform and Gaussian distributed sensor network.