Monte Carlo Simulations for Mine Detection

Abstract

During January, 1998, collaboration between LLNL, UCI and Exdet, Ltd. arranged for the testing and evaluation of a Russian developed antitank mine detection system at the Buried Objects Detection Facility (BODF) located at the Nevada Test Site. BODF is a secured 30-acre facility with approximately 300 live antitank mines that were buried in 1993 and 1994. The burial depths range from a few cm to 15 cm and the various metal- and plastic-case antitank mines each contain 6-12 kg of high explosive. Contractors who have tested their mine detection equipment at BODF include: SAIC, SRI, ERIM, MIT/Lincoln Laboratory and Loral Defense Systems. In addition LLNL researchers have used BODF to test antitank mine detection systems based on: dual-band infrared imaging, hyper-spectral imaging, synthetic aperture impulse radar and micro-impulse radar. In a blind test the Russian operated system obtained the highest score of any technology tested to date at BODF. The system is based on combining information from two separate sensors; one to detect anomalous concentrations of hydrogen and the other to detect if such anomalies also have the correct nitrogen to carbon ratio for high explosives. The detection sensitivity is set by the geometry and type of neutron moderator andmore » filters surrounding the neutron source and detectors. Detection of hydrogen anomalies is a rapid process based on neutron scattering. The handheld instrument on the end of a wand could scan a large area at a rate of 4-5 square meters per minute. Once the hydrogen anomalies were located a second sensor was used to measure the thermal neutron excited gamma-ray spectrum at each hydrogen anomaly to determine whether that location in addition contained high concentrations of nitrogen. The second process was slower, taking up to 5 minutes for each location. The information from both sensors were then examined by the operator and a declaration was made as to whether or not the anomaly was a buried antitank mine. Although the system worked extremely well on all classes of anti-tank mines, the Russian hardware components were inferior to those that are commercially available in the United States, i.e. the NaI(Tl) crystals had significantly higher background levels and poorer resolution than their U.S. counterparts, the electronics appeared to be decades old and the photomultiplier tubes were noisy and lacked gain stabilization circuitry. During the evaluation of this technology, the question that came to mind was: could state-of-the-art sensors and electronics and improved software algorithms lead to a neutron based system that could reliably detect much smaller buried mines; namely antipersonnel mines containing 30-40 grams of high explosive? Our goal in this study was to conduct Monte Carlo simulations to gain better understanding of both phases of the mine detection system and to develop an understanding for the system's overall capabilities and limitations. In addition, we examined possible extensions of this technology to see whether or not state-of-the-art improvements could lead to a reliable anti-personnel mine detection system.« less