Identification of Nuclear Materials from Remote Detection of Characteristic Gamma Rays.

This research effort studies the application of Bayes decision theorem to extending the range of remote identification of nuclear materials. The Finite Element Method was used to develop a radiation transport code which was used to reconstruct a group of three sample sources at distances of 50, 100, 200, 300, 400, and 500 meters. Both the Poisson and multinomial distributions were then used to simulate measured sources in a low count environment at these six ranges. Bayes theorem was applied to the resulting measured sources to test for positive identification. The results show that a low resolution detector can increase the range of remote detection an average of 100 meters when compared with the method of photopeak identification. Bayes theorem is unable, however, to identify sources not contained in the library of known sources. Originator supplied keywords include Gamma Ray Spectra, Bayes Theorem, Remote Detection, Radioactive Materials.