A sequential detector for biochemical release in realistic environments

We develop a sequential detector for the release of a biochemical substance, with potential applications in environmental security. The proposed detector provides online detection of the appearance of a biochemical source in realistic complex environments. To obtain optimal performance, we use an integrated approach combining statistical analysis of measurements given by an array of biochemical sensors with a physical model of the dispersion, amenable to arbitrary geometries and wind turbulence. We first focus on formulating a detector that is applicable in the presence of unknown source parameters (e.g., release time, intensity, and location). We then derive a bound on the expected delay before a false detection in order to select the threshold of the test. For a fixed false-alarm rate, we obtain the detection probability of a release as a function of its location and initial concentration. Numerical examples illustrate the applicability of our methods to real-world scenarios of an urban area and indoor ventilation duct.