¹College of Communication Engineering, Chongqing University, Chongqing, China
²National Laboratory of Analogue Integrated Circuits, Sichuan Institute of Solid-State Circuits, No. 24 Research Institute of China Electronics Technology Group Corporation, Chongqing China
³China Aerodynamics Research and Development Center, Mianyang Sichuan, China

---

Abstract—Fire hazard monitoring and evacuation for building environments is a novel application area for the deployment of wireless sensor networks. In large place such as a marketplace, it is crucial for firefighters to know the fire hazard situation, and decide on how to best tackle the disaster. Fire sensors in the traditional fire hazard monitoring system are isolated in large place, and it results in failure to remain aware of current overall fire conditions. To improve the assessment accuracy of fire situation, a novel wireless sensor network based algorithm is proposed in this paper to detect and localize the origin of fire hazard in large place. Fuzzy artificial neural network is trained by fire signal from the wireless sensor nodes, then determine whether the fire hazard occurred and estimate its situation. The smoke and temperature diffusion models in limited space are adopted to localize the origin of fire hazard. Theory analysis and experimental results show that this fire detection and localization algorithm could more accurately assess fire hazard situation in large place, which is helpful for firefighters to take measures to exterminate fire hazard.

Index Terms—Fire hazard detection, Fire origin localization, Wireless sensor network, Fuzzy artificial neural network

[PDF]

Cite: Zhengzhou Li, Yuanshan Gu, Lan Tang, Hao Luo, Gang Jin, "Fire Hazard Location Algorithm for Large Place Based on Wireless Sensor Network," Journal of Computers vol. 9, no. 3, pp. 711-716, 2014.