Simulation Modeling of Large-Scale Ad-hoc Sensor Networks

In the scenario of a natural catastrophe or a terrorist attack, a large number of self-organizing, low-cost sensor devices can be deployed over the affected area. Each device equipped with its own power source, sensor, processing unit and low-power radio, can be imbued with the intelligence to seek out its neighbours and join in a wireless network spanning the geographic domain. The sensed quantities can then be forwarded to collections points, where the information is aggregated and presented to emergency response teams. We are developing a high-performance framework for the large-scale simulation of wireless ad-hoc networks (SWAN). Our framework is comprised of inter-operating sub-models for terrain, dispersion of hazardous substance, radio propagation, and the actual source code of ad-hoc networking protocols. In this paper, we describe the architecture of this framework and present experiments that confirm its usefulness in the study of routing algorithms.