Bayesian Processing for the Detection of Radioactive Contraband from Uncertain Measurements

With the increase in terrorist activities throughout the world, the need to develop techniques capable of detecting radioactive contraband in a timely manner is a critical requirement. The development of Bayesian processors for the detection of contraband stems from the fact that the posterior distribution is clearly multimodal eliminating the usual Gaussian-based processors. The development of a sequential bootstrap processor for this problem is discussed and shown how it is capable of providing an enhanced signal for eventual detection.