REEF : A reliable and energy efficient framework for Wireless Sensor Networks

This paper presents an efficient and scalable framework called Reliable and Energy Efficient Framework (REEF) for reliable data collection in Wireless Sensor Networks. REEF employs a distributed scheme which enables it to scale to large networks. It partitions the network in to clusters where the node with highest residual energy in a neighborhood become the cluster head. REEF forms a virtual backbone connecting the cluster heads and the sink by selecting some nodes from each cluster as gateway nodes. Sensor nodes report sensed data to their respective cluster heads which use an outlier detection algorithm to detect faulty data. REEF significantly cuts down on energy consumption by ensuring that a large number of sensor nodes can go into a deep sleep mode, in which the radio as well as CPU are switched off, for a major part of their life time. Simulation results demonstrate that REEF uses as low as 50% of the energy for the same accuracy when compared to a recently proposed scheme based on passive listening.