• DocumentCode
    3627050
  • Title

    A Hardware-Oriented Method for Evaluating Complex Polynomials

  • Author

    Milos D. Ercegovac;Jean-Michel Muller

  • Author_Institution
    Computer Science Department, University of California at Los Angeles, Los Angeles, CA 90095, USA, milos@cs.ucla.edu
  • fYear
    2007
  • fDate
    7/1/2007 12:00:00 AM
  • Firstpage
    122
  • Lastpage
    127
  • Abstract
    A hardware-oriented method for evaluating complex polynomials by solving iteratively a system of linear equations is proposed. Its implementation uses a digit-serial iterations on simple and highly regular hardware. The operations involved are defined over the reals. We describe a complex-to-real transform, a complex polynomial evaluation algorithm, the convergence conditions, and a corresponding design and implementation. The latency and the area are estimated for the radix-2 case. The main features of the method are: the latency of about m cycles for an m-bit precision; the cycle time independent of the precision; a design consisting of identical modules; and a digit-serial connections between the modules. The number of modules, each roughly corresponding to serial-parallel multiplier without a carry-propagate adder, is 2(n + I) for evaluating an n-th degree complex polynomial. The method can also be used to compute all successive integer powers of the complex argument with the same latency and a similar implementation cost.
  • Keywords
    "Polynomials","Delay","Hardware","Equations","Transforms","Convergence","Computer science","Iterative algorithms","Algorithm design and analysis","Costs"
  • Publisher
    ieee
  • Conference_Titel
    Application-specific Systems, Architectures and Processors, 2007. ASAP. IEEE International Conf. on
  • ISSN
    1063-6862
  • Print_ISBN
    978-1-4244-1026-2
  • Type

    conf

  • DOI
    10.1109/ASAP.2007.4429968
  • Filename
    4429968