High Performance Algorithm for Signal Decomposition in Gamma Ray Detectors

Period of Performance: 01/01/2005 - 12/31/2005


Phase 2 SBIR

Recipient Firm

Tech-X Corporation
5621 Arapahoe Ave Suite A
Boulder, CO 80303
Principal Investigator
Firm POC


76060S The GRETINA detector, used in nuclear physics research, consists of highly-segmented Ge crystal diodes that produce currents upon impact of gamma rays. The decomposition of these currents into combinations of known signals determines the number, location, and energies of the gamma ray interactions. This decomposition must be done in real time as the experiment runs, which requires fast and efficient algorithms. To address this need, this project will develop signal decomposition software that incorporates singular-value decomposition along with other signal decomposition techniques. In Phase I, the requirements for variable grids for the crystals were examined, Singular Value Decomposition (SVD) was evaluated as a possible algorithm for the signal decomposition, and candidate hybrid algorithms using SVD were identified. Phase II will implement an exact SVD matrix decomposition algorithm that can handle large matrices. Hybrid algorithms for the signal decomposition of highly-segmented gamma ray detectors will be developed, and the algorithms will be applied to the GRETINA experiment. Commercial Applications and Other Benefits as described by the awardee: The software product should provide gamma ray detectors with greater spatial resolution and efficiency than currently available. Such improved detectors should have use not only in nuclear physics but also in astrophysics and nuclear medicine.