The multiple sensor positive detection problem

This paper develops a Bayesian probability formula to infer the presence of targets given multiple, noisy detection reports. This problem is characterized by having only positive reports, because the absence of a target is rarely transmitted to the fusion center. Further characteristics of this problem are spatial uncertainty in detection locations and high false alarm rates. In this paper, we develop a Bayesian probability formula to infer the presence of a target given multiple sensor detections where each sensor has a known spatial uncertainty, probability of detection, and probability of false alarm. Results are shown on simulated data to demonstrate the effectiveness of the algorithm when underlying assumptions are true. In addition to simulated data, we also use a real dataset that involves reported target locations. These results demonstrate the effectiveness of our algorithm in finding probable target locations when all of the detection locations are slightly different.