Smart sensing and sensor data collection on the move for modelling intelligent environments

Abstract

With advent of pervasive computing and considerable acceptance of sensor networks, smart sensing techniques and data collection have been topics of interest. This paper presents a smart sensing and data collection technique from sensor networks using context aware high powered mobile objects within the environment. The paper proposes CAM-R a context aware robot that can move within smart environments sensing new sensor sources and collecting sensory originated data efficiently. Based on these sensed data sources, we propose an extension to context spaces model that builds a virtual model of the intelligent environment. This intelligent environment model built using extended context spaces can be used by number of context aware applications to efficiently query and retrieve data from the sensor network using CAM-R based data collection approach. We also present a prototype implementation of CAM-R built using off-the-shelf hardware and a context based cost function used to compute data collection decisions. We validate our system by implementing the virtual modelling of the intelligent environment based on simulated input obtained from CAM-R and sensors. We also evaluate CAM-Rby simulating and comparing the energy spent by the sensor nodes during data collection process using our proposed approach and traditional fixed sink based approach. © 2008 Springer-Verlag Berlin Heidelberg.