DocumentCode
3348790
Title
Implementation of Linear Algebra Algorithms in FPGA-based Rational Fraction Arithmetic Units
Author
Maslennikow, Oleg ; Ratuszniak, Piotr ; Sergyienko, Anatoli
Author_Institution
Dept. of Electron., Tech. Univ. of Koszalin, Koszalin
fYear
2007
fDate
19-24 Feb. 2007
Firstpage
228
Lastpage
234
Abstract
In this paper, two fixed size processor array architectures, which are destined for realization of several linear algebra algorithms, are proposed. In order to implementation of these architecture in modern FPGA devices, the arithmetic unit (AU) operating in the rational fraction arithmetic is designed, which is adapted to realization in the Xilinx reconfigurable platforms Virtex II or Virtex 4 families. It allows to reduce the hardware complexity of the new AU up to 4,5 times in comparison with similar AUs operating with float-point numbers, without decreasing of AU performance and increasing round off errors.
Keywords
VLSI; field programmable gate arrays; floating point arithmetic; linear algebra; microprocessor chips; reconfigurable architectures; FPGA-based rational fraction arithmetic units; VLSI array processor; Virtex 4; Virtex II; Xilinx reconfigurable platform; field programmable gate array; fixed size processor array architecture; float-point numbers; hardware complexity; linear algebra algorithm; Digital arithmetic; Digital signal processing; Field programmable gate arrays; Gold; Hardware; Linear algebra; Pipeline processing; Problem-solving; Signal processing algorithms; Very large scale integration; FPGA (Field Programmable Gate Array); Linear Algebra Algorithm; Rational Fraction Arithmetic; Rational Fraction Number System; VLSI Array Processor;
fLanguage
English
Publisher
ieee
Conference_Titel
CAD Systems in Microelectronics, 2007. CADSM '07. 9th International Conference - The Experience of Designing and Applications of
Conference_Location
Lviv-Polyana
Print_ISBN
966-533-587-0
Type
conf
DOI
10.1109/CADSM.2007.4297532
Filename
4297532
Link To Document