 |
Center for Nonlinear Studies |
Distributed sensor networks (DSNs) are an emerging technology with potential applications in signal detection, environmental and medical monitoring, surveillance, and tracking. By distributing sensors in a region of space and allowing them to communicate by radio, one can gather information about the entire region and turn it into useful data. In most applications, the sensors must be small, cheap, and they must operate at extremely low power. Since the power is so constrained and sensors are prone to failure, DSNs must use algorithms that minimize false alarms and power consumption. In this talk I describe a new algorithm that enables DSNs to track moving objects in noisy environments where previous algorithms have failed. Based on continuous cellular automata, the algorithm conserves power by requiring sensors to mainly communicate with their nearest neighbors.