Wireless sensor network application for fire hazard detection and monitoring

Abstract

Hazard detection systems are sophisticated tools that can help us detect and prevent environmental disasters. The role of a well designed environmental hazard detection system based on a Wireless Sensor Network (WSN) is to continuously monitor and report the environment's status by sampling relevant physical parameters (e.g. temperature), but at a rate that can be adapted dynamically to the criticality of the current situation, so that precious energy is conserved as much as possible and communication bandwidth is not wasted, both preconditions that need to be met for a scalable WSN application. We have designed and built a small-scale prototype of such a WSN system for fire detection and monitoring based on inexpensive in-house developed wireless sensor nodes. These nodes combine an AVR Butterfly microcontroller demonstration kit with an Xbee wireless Zigbee transceiver. The emphasis of the work reported here is on the software designed for the fire hazard detection application. We discuss the embedded computing strategy developed for the in-field sensor nodes that allows them to adjust their mode of operation (i.e. their sampling and reporting rates) dynamically and in an autonomous manner depending on the area prevailing conditions. We also discuss the functionality of the software running on the central node (PC) that is used to initialize the WSN system, synchronize nodes, monitor their status by maintaining an active registry, adjust parameters at any time, inspect real-time plots of the incoming temperature reports of selected nodes to monitor emerging trends and patterns etc. Several examples of the end-to-end system's use are also presented and discussed. © Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering 2010.