Energy-efficient processing of monitoring queries in wireless sensor networks

Abstract

In recent years, rapid advances in sensing and wireless communication technologies have resulted in the availability of wireless sensor networks. Consisting of several micro-electronic sensor devices, which can collect information of surrounding areas and communicate with each other through radio waves, these networks are widely used in a variety of monitoring applications such as habitat monitoring, environmental observation, and military surveillance. In these applications, small and battery-powered sensor devices are distributed into the environment, periodically sensing and reporting relevant data to the base station to answer continuous monitoring queries submitted by the users. While these applications require highly accurate query results and long-term performance of the network, wireless sensor networks suffer from the energy limitation that do not appear in traditionally wired computing environments.

In this dissertation, we study the processing of three representative types of monitoring queries so that the energy at sensor devices is used as efficient as possible. First of all, top-k monitoring queries are investigated due to their usefulness in many practical monitoring situations. A query of this type continuously returns a list of k ordered sensor nodes with the highest (or lowest) sensor readings to the base station. To process these queries, a well-known approach is to install a filter at each sensor node to avoid unnecessary transmissions of the sensor readings. Based on this approach, we propose a new top-k monitoring method, named Distributed Adaptive Filter-based Monitoring. In this method, we first propose a new query reevaluation algorithm that works distributedly in the network to reduce the communication cost of sending probe messages. Then, we present an adaptive filter updating algorithm which is based on predicted benefits to lower down the transmission cost of sending updated filters to the sensor nodes. Experimental results on rea...