Numerical Integration with Graphical Processing Unit for QKD Simulation
Master's thesis Sep 2012-May 2014
AIR FORCE INSTITUTE OF TECHNOLOGY WRIGHT-PATTERSON AFB OH GRADUATE SCHOOL OF ENGINEERING AND MANAGEMENT
Pagination or Media Count:
The Air Force Institute of Techology AFIT is developing a simulation framework to model a wide variety of existing and proposed Quantum Key Distribution QKD systems. This research investigates using graphical processing unit GPU technology to more efficiently integrate optical pulses modeled within this framework. The goal is to reduce the simulation execution time. A GPU algorithm is presented for performing numerical integration of optical pulses described by Gaussian curves to improve pulse energy and power calculations. In order to measure the performance of the algorithm a optimal timing method is needed. A timer using Comute Unified Device Architecture CUDA events is selected over a Windows system application programming interface API timer. The problem sizes studied produce speedups greater than 60x on the NVIDIA Tesla C2075 compared to the Intel i7-3610QM CPU.
- Theoretical Mathematics
- Computer Programming and Software