Previous [ 1] [ 2] [ 3] [ 4] [ 5] [ 6] [ 7] [ 8] [ 9] [ 10] [ 11] [ 12] [ 13] [ 14] [ 15] [ 16] [ 17] [ 18] [ 19] [ 20] [ 21] [ 22] [ 23] [ 24]


Journal of Information Science and Engineering, Vol. 27 No. 1, pp. 51-64 (January 2011)

Efficient Emergency Rescue Navigation with Wireless Sensor Networks*

State Key Laboratory for Novel Software Technology
Nanjing University
Nanjing, Jiangsu 210093, P.R. China
E-mail:; {zad@dislab.; wuxb@; gchen@};

Recently, many applications in wireless sensor networks (WSNs) have been discussed. Navigation with WSNs is among the most heated debated ones. Previous navigation algorithms attempt to find safe and efficient escape paths for individuals under various environmental dynamics but ignore possible congestion caused by the individuals rushing for the exits. Moreover, most previous works have overlooked the fact that the emergency rescue force can take actions strategically in order to save people out of danger. We propose an efficient Emergency Rescue Navigation strategy (ERN) by treating WSNs as navigation infrastructure. Our approach takes both pedestrian congestion and rescue force flexibility into account. A directed graph is used to model the emergency regions. Humans movements are regarded as network flows on the graph. By calculating the maximum flow and minimum cut on the graph, the system can provide firemen rescue commands to eliminate key dangerous areas, which may significantly reduce congestion and save trapped people. We have performed extensive simulations under dynamic environments to evaluate the effectiveness and response time of our work. Simulation results show that, with our strategy, people in emergency are evacuated much faster and less congestion is observed.

Keywords: wireless sensor networks, cyber-physical system, navigation, maximum network flow, dinic algorithm

Full Text () Retrieve PDF document (201101_04.pdf)

Received October 30, 2009; revised February 28, 2010; accepted June 24, 2010.
Communicated by Ren-Hung Hwang, Chung-Ming Huang, Cho-Li Wang, and Sheng-Tzong Cheng.
* The work is partly supported by China NSF grants (60721002, 60825205, 60903179, 61073152), and China 973 project (2006CB303000).
+ Correspondence author.