Sensor network placement for maximizing detection of vehicle tracks and minimizing disjoint coverage areas

The effective placement of intelligence, surveillance and reconnaissance (ISR) assets in a sensor network for optimal situational awareness is an important area of military operations research. The sensor network placement optimization problem is a type of assignment problem where possible sensor locations can be assigned sensors of different types (or no sensor). We apply a genetic algorithm to the sensor network node assignment problem, with the goal of optimizing for barrier, rather than blanket, coverage. This shifts the measure of effectiveness of the network from the amount of area coverage to continuity of coverage and the number of incoming targets detected. To address this problem, we simulate vehicle tracks crossing a nominal area of interest. We then apply the Non-dominated Sorting Genetic Algorithm II (NSGA II) to determine the placement of sensors with the following objectives: to maximize the number of detected vehicle tracks, to minimize the number of discontinuities in coverage, and to minimize total sensor network cost. In this paper, we describe the sensor network node assignment problem, and our specific implementation of NSGA II. We then present a specific test scenario and a preliminary example of the generated non-dominated front including various sensor network configurations. Finally, we outline future work.