A probabilistic model for the deployment of sensors

Coverage is an important optimization objective in sensor deployment problems. This paper addresses the issue of covering a set of target points in an area with a finite set of sensors. A probabilistic model is proposed which takes in account the detection probabilities of the sensing devices which may decay with distance, environmental conditions, and hardware configuration. The objective is to deploy sensors so that the distribution of the sensors meets the probability of detection requirements while minimizing costs. The expected points to cover and the deployment points are assumed to be stationary and known a priori. A probabilistic coverage matrix is defined and the deployment is optimized using a genetic algorithm. Our experimental results verify that the proposed probabilistic sensor deployment model finds more efficient solutions requiring fewer sensors compared to other deployment schemes.