Bayesian Quickest Detection of Propagating Spatial Events

Rapid detection of spatial events that propagate across a sensor network is wide interest in many modern applications. In particular, communications, radar, IoT, environmental monitoring, and biosurveillance, we may observe propagating fields or particles. this paper, propose Bayesian sequential single multiple change-point procedures for the rapid such phenomena. Using dynamic programming framework derive structure optimal single-event quickest procedure, which minimizes average delay (ADD) subject to false alarm probability upper bound. The multi-sensor system configuration arbitrary sensors be mobile. rare event regime, procedure converges more practical threshold test on posterior change point. A convenient recursive computation derived by using propagation characteristics event. ADD analyzed asymptotic specific analysis conducted setting detecting random Gaussian signals affected path loss. Then, show how proposed easy extend parallel hypothesis testing setting. method provides strict discovery rate (FDR) control proposed. simulation section, it demonstrated exploiting properties decreases compared do not utilize information, even under model mismatch.