Forest Fire Modeling and Early Detection using Wireless Sensor Networks

Forest fires cost millions of dollars in damages and claim many human lives every year. Apart from preventive measures, early detection and suppression of fires is the only way to minimize the damages and casualties. We present the design and evaluation of a wireless sensor network for early detection of forest fires. We first present the key aspects in modeling forest fires. We do this by analyzing the Fire Weather Index (FWI) System, and show how its different components can be used in designing efficient fire detection systems. The FWI System is one of the most comprehensive forest fire danger rating systems in North America, and it is backed by several decades of forestry research. The analysis of the FWI System could be of interest in its own right to researchers working in the sensor network area and to sensor manufacturers who can optimize the communication and sensing modules of their products to better fit forest fire detection systems. Then, we model the forest fire detection problem as a coverage problem in wireless sensor networks, and we present a distributed algorithm to solve it. In addition, we show how our algorithm can achieve various coverage degrees at different subareas of the forest, which can be used to provide unequal monitoring quality of forest zones. Unequal monitoring is important to protect residential and industrial neighborhoods close to forests. Finally, we present a simple data aggregation scheme based on the FWI System. This data aggregation scheme significantly prolongs the network lifetime, because it only delivers the data that is of interest to the application. We validate several aspects of our design using simulation.