A Cluster-Based Model for Object Detecting with Wireless Sensor Networks

A system model for wireless sensor networks is presented for efficiently realizing objective detecting. By self-organizing mechanism based on the maximum number of neighboring nodes, a data-gathering algorithm is proposed that partitions network nodes deployed in the detected region into several clusters for the fusion of the sensed objective data. Each cluster consists of a head node and several member nodes, taking responsibility for the fusion and transmitting to base station (BS) of sensed data through multihop communication and being responsible for sensing data respectively. In the process of data gathering, the energy efficiency-aware mechanism insure each node stand for one of four conditions including sleep, passive, test, and active, which are determined by the neighbor node number threshold (NT) and the average data loss rate (DL). Simulations prove that the proposed scheme is valid through put ratio, mean time delay, and packet loss ratio.