Wireless sensor networks are being increasingly used to develop high-tech advance warning systems for natural and human-induced disasters.

Now a team of final-year BTech students of Electronics and Communication Engineering, Ramith Ramesh, Veena Krishnan and Sminu Soman from the Amrita School of Engineering, Bangalore, are working on An Innovative Fire Escape System Using Wireless Sensor Networks.

Their project is funded by the Karnataka State Council for Science and Technology and guided by R. V. Sanjika Devi, Assistant Professor, Department of Electronics and Communication Engineering and N. S. Sreekanth, Senior Technical Officer, ILCG, C-DAC Bangalore.

The students previously completed some parts of the project when interning at C-DAC Bangalore, during the previous years of their BTech studies.

“We proposed the use of a wireless sensor network to detect fire, set-off a fire alarm and identify safe evacuation routes from the building,” the students shared.

They were motivated, in part, by a massive fire that broke out in February 2010 in Bangalore’s Carlton Towers, a multi-storied commercial complex. What the students learned is that escape routes are not always so evident, when such disasters happen, and this worsens the casualties.

The team used Dijkstra’s algorithm to propose their solution.

“Dijkstra’s algorithm is a graph search algorithm that solves for the single-source shortest path problem for a graph with non-negative edge path costs, producing a shortest path tree. This algorithm is often used in routing and as a subroutine in other graph algorithms,” the students explained.

Using this as the basis, the students designed a short route identification algorithm that could enable those trapped in a building to easily locate the fire escape routes.

The team also proposed deployment of temperature sensor nodes in a building to detect fire accidents.

“The weights of edges present in the graph structure of the building change based on the temperature measured at the different places, in case of a fire breakout in the building,” the students clarified.

“Hence after the addition of weights to the graph by applying the Dijkstra’s algorithm, we can get the shortest and safest route to the exits. Then, the obtained information about the safe evacuation routes can be communicated to the occupants of the building,” they added.

Of course, all of this assumes that the system itself is not damaged and remains functional when the fire breaks out.

The students plan to build on their work by making a mobile-based application that can guide people to safety from their current positions in the building, during fire accidents.
We wish them all the best!

August 15, 2013
School of Engineering, Bangalore