Title :
Multiply/Add tradeoffs in length-2nFFT algorithms
Author :
Heideman, M. ; Burrus, C. Sidney
Author_Institution :
Rice University, Houston, Texas
Abstract :
The multiplicative complexity of the length-2ndiscrete Fourier Transform is derived. A constructive approach is followed which describes the set of algorithms that realize the minimum number of multiplications. The operation counts of minimum multiply algorithms are compared to other FFT algorithms. The basic method is extended to derive the multiplicative complexity of length-pnDFFs where p is an odd prime number.
Keywords :
Algorithm design and analysis; Arithmetic; Computational complexity; Discrete Fourier transforms; NASA; Polynomials; Signal processing algorithms; Upper bound;
Conference_Titel :
Acoustics, Speech, and Signal Processing, IEEE International Conference on ICASSP '85.
DOI :
10.1109/ICASSP.1985.1168331