Sensor networks for the detection and tracking of radiation and other threats in cities
This paper presents results from experiments, mathematical analysis, and simulations of a network of static and mobile sensors for detecting threats on city streets and in open areas such as parks. The paper focuses on the detection of nuclear radiation threats and shows how the analysis can be extended to other classes of threat. The paper evaluates algorithms that integrate methods of parametric and Bayesian statistics. A pure Bayesian approach is difficult because obtaining prior distributions on the large number of parameters is challenging. The results of analyses and simulations are compared against measurements made on a reduced scale testbed. A survey of background radiation in the city of Sacramento is used to quantify the efficacy of police patrols to detect threats. The paper also presents algorithms that optimize network parameters such as sensor placement.
© 2011 ACM. We would like to thank Bryan Cutler, Steve Foote, Andy Gooden, and Bozhil Makaveev of IOS, Pasadena for their assistance with performing the experiments described in the paper, and to thank IOS for the loan of their equipment. We also wish to thank David Campbell, David Trombino from the Lawrence Livermore National Laboratory for providing radiation measurements and analysis tools. Special thanks to Karl Nelson and Simon Labov from the Lawrence Livermore National Laboratory for their close guidance on analyzing radiation measurements. The Sacramento measurement work was performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344. Our research was supported in part by the National Science Foundation Cyber Physical Systems program.