An Energy-Efficient Sequence-Aware Top-k Monitoring Scheme in Wireless Sensor Networks

Abstract

We focus on top- k monitoring in wireless sensor networks and propose a novel sequence-aware top- k monitoring algorithm called SAT. Top- k monitoring is important to many applications of sensor networks. Conventional top- k monitoring algorithms install a filter at each sensor node and suppress unnecessary sensor updates. However, they have some drawbacks such as the fact that the sensor nodes consume energy extremely to probe sensor reading or to update filters. Our basic idea is to collect readings sequentially by their values. First, sequence-aware data collection is investigated to make sensor nodes to determine their orders for data gathering phase. Next, sensor nodes transmit their sensor readings sequentially to the base station. When the base station collects k-readings, it broadcasts a simple message to stop data gathering phase. Therefore, SAT may minimize the communication cost for processing top- k queries. Moreover, we expand our approach to a cluster-based top- k monitoring to filter out false positives in hierarchical levels. In order to show the superiority of our top- k monitoring approach, we simulate its performance with the conventional filter-based top- k monitoring algorithm. In the results, our approach reduces communication overhead and prolongs the network lifetime largely.