Capacity Analysis for Flat and Clustered Wireless Sensor Networks

Wireless sensor networks are often divided into clusters to gain high-performance and prolong network lifetime. In this paper we analyze the network capacity of two different architectures, named flat sensor network and clustered sensor network. For each network, we first define the network model and then analyze the network capacity and the factors that affect the capacity. Numerical results suggest that clustering can significantly improve the network capacity. We also found that the capacity of a clustered network is saturated after the number of clusters is above a certain value. For clustered network architecture, we further study the capacity difference with or without the distributed coordination function. It is found that there is a critical value of the number of cluster heads beyond which the network capacity of two clustered models approaches the same. Results from this research will provide guidance in deploying high-performance wireless sensor networks.