Energy Aware Queries in Sensor Networks

Abstract

Energy is very important in sensor networks because sensor motes pow- ered by batteries with limited capacity. Energy e cient operating systems, protocols and applications are emerging to overcome energy limitations. Sen- sor network database abstractions provide user friendly interface to access sensor data. User unawareness of limitations in sensor networks may lead them to make wrong decision when querying data from sensor network databases. User awareness about energy or lifetime of sensor network will improve the e ective usage of sensor network. This research presents a novel approach to acquire valuable information about remaining power or lifetime from sensor networks. Remaining energy is very important when executing long running queries against sensor network databases. Considering the importance of energy related information in sen- sor network database applications we introduced Energy Aware Queries. This new query type provides meaningful information about remaining energy of sensor motes and whole sensor network. In this approach energy is referred to the lifetime of motes while performing some task. The underline concept of the energy aware query is energy estimation. Query results are the es- timation outputs and it is pre estimation of energy before executing actual queries on the network. Cost based energy model is utilized for estimation process because of the limitation of target software and hardware platforms. A detailed and accurate energy model guaranteed the accuracy of the estima- tions and query results. Energy aware queries are implemented on TikiriDB sensor network database abstraction and energy model is implemented on Micaz hardware platform.