International Journal of Advanced Research in Computer and Communication Engineering

A monthly peer-reviewed online and print journal

ISSN Online 2278-1021
ISSN Print 2319-5940

Since 2012

Abstract: The fused floating point units are advantageous for various Digital Signal Processing (DSP) applications such as butterfly operations of Fast Fourier Transform (FFT) and Discrete Cosine Transform (DCT) and complex multiplication. In this thesis, two floating point fused units are proposed: 1) Floating point Fused dot product unit, 2) Floating point fused add subtract unit. Fused dot product unit performs the multiplication and addition of two pairs of operands. Fused add subtract unit perform the addition and subtraction of same operands simultaneously. In order to enhance the performance for the fused dot product unit, DADDA multiplication algorithm is applied. FDP unit with DADDA Multiplier improved the speed by 61.1 % compared to the FDP unit with Booth multiplier. The fused dot product unit and fused add subtract unit are implemented for single precision and synthesized in 90nm technology.

Keywords: Fast Fourier Transform (FFT), Discrete Cosine Transform (DCT), Digital Signal Processing (DSP)

PDF | DOI: 10.17148/IJARCCE.2019.8551

Open chat
Chat with IJARCCE