• DocumentCode
    259012
  • Title

    A Low Latency High-Throughput Elementary Function Generator Based on Enhanced Double Rotation CORDIC

  • Author

    Lingxuan Deng ; Junshe An

  • Author_Institution
    Nat. Space Sci. Center, Beijing, China
  • fYear
    2014
  • fDate
    26-27 July 2014
  • Firstpage
    125
  • Lastpage
    130
  • Abstract
    Coordinate Rotation Digital Computer (CORDIC) algorithm is well known and widely used in many computer systems due to its simplicity and the variety of functions it provides. Although CORDIC is known to be able to work in several modes and coordinate to provide elementary functions including trigonometric and transcendental functions, most proposed CORDIC enhancement and application were only limited to a single or only a few modes, and that does not help to build a CORDIC coprocessor that works in all modes. This paper presents an enhanced version of double rotation method which is a scaling-free and redundant arithmetic based CORDIC variation. The proposed algorithm has been implemented and it is shown that the new architectures have great advantage on both FPGA and ASIC platforms and are extremely attractive for high precision and high performance computing.
  • Keywords
    coprocessors; digital arithmetic; function generators; ASIC platforms; CORDIC coprocessor; FPGA platforms; computer systems; coordinate rotation digital computer algorithm; enhanced double rotation CORDIC method; high performance computing; low latency high-throughput elementary function generator; redundant arithmetic based CORDIC variation; scaling-free arithmetic based CORDIC variation; transcendental functions; trigonometric functions; Adders; Computer architecture; Coprocessors; Field programmable gate arrays; Hardware; Vectors; ASIC; CORDIC; FPGA; coprocessor; double rotation; elementary function; redundant arithmetic; signed-digit;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Computer Applications and Communications (SCAC), 2014 IEEE Symposium on
  • Conference_Location
    Weihai
  • Type

    conf

  • DOI
    10.1109/SCAC.2014.33
  • Filename
    6913181