Title :
A root of less evil [digital signal processing]
Author :
Allie, Mark ; Lyons, Richard
Author_Institution :
Dept. of Electr. & Comput. Eng., Wisconsin Univ., Madison, WI, USA
fDate :
3/1/2005 12:00:00 AM
Abstract :
This work discusses several methods for estimating the square root of a number, with particular focus on high-speed techniques for approximating the square root of a single value as well as the square root of a sum of squares for quadrature (I/Q) vector magnitude estimation. The paper presents the Newton-Raphson inverse method and the nonlinear IIR filter method for estimating the square root of a single value using iterative methods. It also presents the binary-shift magnitude estimation method and the equiripple-error magnitude estimation method for estimating the magnitude of a complex number.
Keywords :
IIR filters; Newton-Raphson method; fixed point arithmetic; nonlinear filters; complex vector magnitude approximation algorithm; fixed-point fractional binary arithmetic; high-throughput algorithm; iterative Newton-Raphson inverse square root method; nonlinear IIR filter (NIIRF) square root method; Arithmetic; Demodulation; Digital signal processing; Fast Fourier transforms; Gain control; Graphics; Iterative methods; Root mean square; Signal processing algorithms; Table lookup;
Journal_Title :
Signal Processing Magazine, IEEE
DOI :
10.1109/MSP.2005.1406500
