Emulating Poisson Emissions with a Pulsed Laser
University of Delaware
The problem of detecting hazardous radioactive materials is receiving considerable attention due to emerging threats of clandestine nuclear attacks. Protection against these threats requires a multilayered defense approach, combining border security through cargo screening with overseas nonproliferation enforcement. Current technology is insufficient for applications that involve searching or mapping in unstructured environments. Instead of using large expensive radiographic imaging sensors at controlled security stations, many applications require the use of multiple less sophisticated sensors. These sensor networks can be mobile in general and can adaptively reconfigure themselves to collect more useful measurements (e.g. moving sensors closer to inspect a potential source). Determining how to optimally control sensor networks and process collected information for quick and confident detection is a difficult problem that has driven recent theoretical developments. And although the physics behind emission and sensing radiation is well understood and our models typically work well, there is still a need for experimental testing of newly developed theory. Not only can experimentation validate our expectations of system behavior, but it can also provide valuable insight for improving our models and control strategies. However, using actual fissile materials for experiments would be hazardous and impractical. The aim of this research is to develop a novel system that emulates the process of nuclear emission to use as an innocuous proxy for experimental testing. The device developed uses a modulated laser pointer to produce sporadic pulses of light and a rotating mirror that reflects each pulse in a random direction, mimicking the way in which gamma rays are emitted from a radiation source. The performance of this emulation is compared to that of a weakly radioactive source through side by side experiments. First, the pulse frequency distribution is shown to conform to the desired Poisson process, which is the underlying process used to model weak radiation detection. Next, the distance-dissipation effect is verified through static tests at various distances. The device is then applied to two simple detection regimes: static fixed interval detection and detection of a source moving in a straight line with a combination of static and mobile sensors. The results show a strong agreement with simulations and thus, the device is concluded to be a viable proxy for experimentation. The pros and cons of using the device are discussed and careful attention is given to analyzing the conditions under which it can be used.