Abstract
Gaussian random number generators (GRNGs) are an important component in parallel Monte Carlo simulations using FPGAs, where tens or hundreds of high-quality Gaussian samples must be generated per cycle using very few logic resources. This article describes the Table-Hadamard generator, which is a GRNG designed to generate multiple streams of random numbers in parallel. It uses discrete table distributions to generate pseudo-Gaussian base samples, then a parallel Hadamard transform to efficiently apply the central limit theorem. When generating 64 output samples, the Table-Hadamard requires just 130 slices per generated sample, which is a third of the resources needed by the next best technique, while still providing higher statistical quality.
- Amirhossein Alimohammad, Saeed Fouladi Fard, Bruce F. Cockburn, and Christian Schlegel. 2008. A compact and accurate Gaussian variate generator. IEEE Transactions on Very Large Scale Integration (VLSI) Systems 16, 5, 517--527. Google Scholar
Digital Library
- Ray J. Andraka and Richard M. Phelps. 1998. An FPGA based processor yields a real time high fidelity radar environment simulator. In Proceedings of the Conference on Military and Aerospace Applications of Programmable Devices and Technologies.Google Scholar
- Andrew C. Berry. 1941. The accuracy of the Gaussian approximation to the sum of independent variates. Transactions of the American Mathematical Society 49, 1, 122--136.Google Scholar
Cross Ref
- George E. P. Box and Mervin E. Muller. 1958. A note on the generation of random normal deviates. Annals of Mathematical Statistics 29, 2, 610--611.Google Scholar
Cross Ref
- Ray C. C. Cheung, Dong-U Lee, Wayne Luk, and John D. Villasenor. 2007. Hardware generation of arbitrary random number distributions from uniform distributions via the inverson method. IEEE Transactions on Very Large Scale Integration (VLSI) Systems 15, 8, 952--962. Google Scholar
Digital Library
- Ishaan L. Dalal and Deian Stefan. 2008. A hardware framework for the fast generation of multiple long-period random number streams. In Proceedings of the 16th International ACM/SIGDA Symposium on Field Programmable Gate Arrays (FPGA’08). 245--254. Google Scholar
Digital Library
- Dong-U Lee, John D. Villasenor, Wayne Luk, and Philip H. W. Leong. 2006. A hardware Gaussian noise generator using the Box-Muller method and its error analysis. IEEE Transactions on Computers 55, 6, 659--671. Google Scholar
Digital Library
- Dong-U Lee, Wayne Luk, John D. Villasenor, Guanglie Zhang, and Philip H. W. Leong. 2005. A hardware Gaussian noise generator using the Wallace method. IEEE Transactions on Very Large Scale Integration (VLSI) Systems 13, 8, 911--920. Google Scholar
Digital Library
- Jamshaid Sarwar Malik, Jameel Nawaz Malik, Ahmed Hemani, and Nasir D. Gohar. 2011. Generating high tail accuracy Gaussian random numbers in hardware using central limit theorem. In Proceedings of the 2011 IEEE/IFIP 19th International Conference on VLSI and System-on-Chip (VLSI-SoC’11). 60--65.Google Scholar
- Sean O’Connor. 2006. The Hadamard Transform Fact Sheet. Retrieved August 24, 2015, from http://www.dsprelated.com/showmessage/65499/1.php.Google Scholar
- David B. Thomas and Wayne Luk. 2007a. High quality uniform random number generation using LUT optimised state-transition matrices. Journal of VLSI Signal Processing 47, 1, 77--92. Google Scholar
Digital Library
- David B. Thomas and Wayne Luk. 2007b. Non-uniform random number generation through piecewise linear approximations. IET Computers and Digital Techniques 1, 4, 312--321.Google Scholar
Cross Ref
- David B. Thomas and Wayne Luk. 2010a. FPGA-optimised uniform random number generators using LUTs and shift registers. In Proceedings of the 2010 International Conference on Field Programmable Logic and Applications (FPL’10). 77--82. Google Scholar
Digital Library
- David B. Thomas and Wayne Luk. 2010b. An FPGA-specific algorithm for direct generation of multi-variate gaussian random numbers. In Proceedings of the 21st IEEE International Conference on Application-Specific Systems, Architectures, and Processors. 208--215.Google Scholar
- Anson H. T. Tse, David B. Thomas, Kuen H. Tsoi, and Wayne Luk. 2011. Efficient reconfigurable design for pricing Asian options. ACM SIGARCH Computer Architecture News 38, 4, 14--20. DOI:http://dx.doi.org/10.1145/1926367.1926371 Google Scholar
Digital Library
- Michael J. Wichura. 1988. Algorithm AS 241: The percentage points of the normal distribution. Applied Statistics 37, 3, 477--484.Google Scholar
Cross Ref
- Guanglie L. Zhang, Philip H. W. Leong, Dong-U Lee, John D. Villasenor, Ray C. C. Cheung, and Wayne Luk. 2005. Ziggurat-based hardware Gaussian random number generator. In Proceedings of the International Conference on Field Programmable Logic and Applications. IEEE, Los Alamitos, CA, 275--280.Google Scholar
Index Terms
The Table-Hadamard GRNG: An Area-Efficient FPGA Gaussian Random Number Generator
Recommendations
Recycling random bits in parallel
HICSS '95: Proceedings of the 28th Hawaii International Conference on System SciencesShows that r pseudo-random bits can be obtained by concatenating t blocks of r/t pseudo-random bits, where the blocks are generated in parallel. This can be considered as a parallel version of R. Impagliazzo and D. Zuckerman's (1989) method of recycling ...
Parallel Generation of Gaussian Random Numbers Using the Table-Hadamard Transform
FCCM '13: Proceedings of the 2013 IEEE 21st Annual International Symposium on Field-Programmable Custom Computing MachinesGaussian Random Number Generators (GRNGs) are an important component in parallel Monte-Carlo simulations using FPGAs, where tens or hundreds of high-quality Gaussian samples must be generated per cycle using very few logic resources. This paper ...
Testing parallel random number generators
Monte Carlo computations are considered easy to parallelize. However, the results can be adversely affected by defects in the parallel pseudorandom number generator used. A parallel pseudorandom number generator must be tested for two types of ...






Comments