BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:ARC Preview: Word-length Optimization and Error Analysis of a Mult
 ivariate Gaussian Random Number Generator - Paul Saiprasert (Imperial Coll
 ege)
DTSTART:20090305T154500Z
DTEND:20090305T161500Z
UID:TALK16586@talks.cam.ac.uk
CONTACT:Dr George A Constantinides
DESCRIPTION:Monte Carlo simulation is one of the most widely used techniqu
 es for computationally intensive simulations in mathematical analysis and 
 modeling. A multivariate Gaussian random number generator is one of the ma
 in building blocks of such a system. Field Programmable Gate Arrays (FPGAs
 ) are gaining increased popularity as an alternative means to the traditio
 nal general purpose processors targeting the acceleration of the computati
 onally expensive random number generator block. This paper presents a nove
 l approach for mapping a multivariate Gaussian random number generator ont
 o an FPGA by automatically optimizing the computational path with respect 
 to the resource usage. The proposed approach is based on the Eigenvalue de
 composition algorithm which decomposes the design into computational paths
  with different precision requirements. Moreover\, an error analysis on th
 e impact of the error due to truncation is performed in order to provide u
 pper bounds of the error inserted into the system. The proposed methodolog
 y optimises the usage of the available FPGA resources leading to area effi
 cient designs without any significant penalty on the overall performance. 
 Experimental results reveal that the hardware resource usage on an FPGA is
  reduced by a factor of two in comparison to current methods.
LOCATION:Mahanakorn Laboratory\, EEE
END:VEVENT
END:VCALENDAR