Emergency evacuation is the most important task at a site, struck by a disaster and the task is best accomplished when an up-to-date map of the disaster site is readily available. This paper addresses the problem of generating a disaster site map of a building on fire using wireless sensor networks (WSN) which is dynamically updated to reflect the current state of disaster. Our technique assumes that the building structure is known apriori and makes use of localization and interpolation to generate hazard values at fine granularity. Interpolation allows us to obtain hazard values at places where there is no sensors and localization (using a selected UWB sensors) provides the approximate locations information of sensor measuring hazard values. Using realistic fire values, through simulation, we compare the generated map with the actual hazard values generated map and show that our technique achieves almost 90% accuracy.
P. S. Mohan Vaishnav, K. Haneesh, S., Srikanth, C. Sai, Koundinya, C., and Subhasri Duttagupta, “Disaster Site Map Generation Using Wireless Sensor Networks”, in Inventive Computation Technologies, Cham, 2020.