Continuous phenomena boundary detection and tracking in wireless sensor networks through optimized monitoring

Abstract

To track and detect continuous natured objects in wireless sensor networks, large number of sensor nodes are involved. These continuous objects like bio-chemical diffusions, forest fires, oil spills usually spread over larger area. Phenomena sensing nodes need to communicate with each other for information exchange and also send sensing information to sink, possibly by passing through many intermediate nodes. The process consumes a lot of communication overhead and waste of sensors scarce energy. In this paper, we proposed an optimized approach to reduce number of communications and report data size to the sink. Sensors locally broadcast their sensing status to one-hop neighbors, and suppress any transmission that contains any trivial data. Intermediate nodes further suppress the report data by applying data aggregation and filtering. Simulation results show a significant decrease in number of communications and report data size.