Performance Analysis of a General Tree Structure for Target Detection in Wireless Sensor Networks
Hadi Varaee, Ghasem Mirjalily, Ali Akbar Tadaion · 2009
One of the basic tasks of a wireless sensor network is target detection, and one of the main issues must be addressed in a detection sensor network is its active topology. Here, we assume a simple hierarchical topology, where sensors that are close to each other will form a cluster and each cluster has its own cluster head, which serves as the local fusion center. Based on decisions received from sensors located within a specific cluster, the corresponding cluster head will make a decision about target presence/absence within that sub-region. The decisions from cluster heads will be further transmitted to the central fusion node to inform it if there is a target or event in specific sub-regions. In this paper, we propose a general tree structure to fuse decisions within a cluster. The aim is to investigate the effects of using different tree topologies in a cluster on the detection performance and energy consumption in a detection sensor network. We will find a tradeoff between high performance and low energy consumption. For this purpose, a function named energy-error has been defined. By finding the minimum value of this function, it is possible to achieve a tradeoff between performance and energy consumption.