Three-Dimensional Localization of Low Activity Gamma-Ray Sources in Real-Time Scenarios


Radioactive source localization plays an important role in tracking radiation threats in homeland security tasks. Its real-time application requires computationally efficient and reasonably accurate algorithms even with limited data to support detection with minimum uncertainty. This paper describes a statistic-based grid-refinement method for backtracing the position of a gamma-ray source in a three-dimensional domain in real-time. The developed algorithm used measurements from various known detector positions to localize the source. This algorithm is based on an inverse-square relationship between source intensity at a detector and the distance from the source to the detector. The domain discretization was developed and implemented in MATLAB. The algorithm was tested and verified from simulation results of an ideal case of a point source in non-attenuating medium. Subsequently, an experimental validation of the algorithm was performed to determine the suitability of deploying this scheme in real-time scenarios. Using the measurements from five known detector positions and for a measurement time of 3 min, the source position was estimated with an accuracy of approximately 53 cm. The accuracy improved and stabilized to approximately 25 cm for higher measurement times. It was concluded that the error in source localization was primarily due to detection uncertainties. In verification and experimental validation of the algorithm, the distance between 137Cs source and any detector position was between 0.84 m and 1.77 m. The results were also compared with the least squares method. Since the discretization algorithm was validated with a weak source, it is expected that it can localize the source of higher activity in real-time. It is believed that for the same physical placement of source and detectors, a source of approximate activity 0.61-0.92 mCi can be localized in real-time with 1 s of measurement time and same accuracy. The accuracy and computational efficiency of the developed scheme make this algorithm a suitable candidate for its deployment in real-time localization of radioactive sources.


Mining and Nuclear Engineering


The work presented in this paper is supported by U.S. NRC under the Award Number NRC-HQ-11-G-38-0008.

Keywords and Phrases

Algorithms; Gamma rays; Inverse problems; Least squares approximations; MATLAB; Nuclear fuels; Radiation protection; Radioactive prospecting; Radioactivity; Scintillation counters; Technology transfer; Border protection; Domain discretization; Nuclear non-proliferation; Radiation detection; Radiation source; Computational efficiency; Nuclear nonproliferation; Radiation source localization; Scintillation detector

International Standard Serial Number (ISSN)


Document Type

Article - Journal

Document Version


File Type





© 2016 Elsevier, All rights reserved.

Publication Date

01 Mar 2016