Experimental Analysis of WSN based Solution for Early Forest Fire Detection

Abstract

Forest fire is a burning issue worldwide. It destroys the green belt, increases air pollution, damages the natural habitat for trees and animals, and even sometimes it causes death to the forest fire fighters. Economically and environmentally it is responsible for huge losses to the country. The satellite-based classical system lags continuous monitoring, requires complex processing and the information does not reach the ground in real-time. Hence, there is a need for a 24/7 forest fire monitoring system that will generate alerts in near real-time. In this paper, we discuss a WSN based solution with visualization and alert generation schemes. The system has been developed and field-tested. The sensors in the system monitor environmental parameters during an outbreak of a forest fire. Sensor data gets processed and upon detection of the outbreak of a fire, alerts are issued to the concerned forest officials. Experimental analysis of sensor node behaviour during a fire in a test set-up has been discussed in detail. Near real-time alerting and prompt response to the alerts might help to mitigate the impact of forest fires such as loss of human lives, loss of large green belts, areas of vegetation, economic loss, and environmental pollution.