Idle-listening Reduction for Data Aggregation in Distributed Sensor Networks

For existing sensor systems, the power consumption of radio for idle listening is in the same order with that for transmitting. In order to prolong the lifetime of sensor networks, we propose a reliable and energy-efficient data aggregation scheme (Reed) to reduce idle listening time. In this application scenario, each sensor node periodically detects the environment and is allowed to send one and only one data packet in each sampling period T . Based on this property, the idle listening time can be significantly reduced by node scheduling, while the network routing topology is maintained. Three techniques are used in Reed: First, the sensor nodes in the network are divided into two types: dominating nodes and non-dominating nodes. Non-dominating nodes sleep for most of the time and only need to awake to broadcast sensing data periodically. Dominating nodes maintain network topology and aggregate sensing data to the sink. Second, role rotation is used to balance the energy consumption of dominating nodes and non-dominating nodes. Third, for the dominating nodes, the idle listening time can be reduced further by predicting the incoming time of packets from neighbors. Analysis and simulation results show that Reed can achieve high reliability and can significantly reduce the energy consumption. It was shown that when the sampling period T = 10s, the energy consumption of Reed is about 70 times less than the network in which all