TOFAST -Fast Direct and Inverse Discrete Fourier Transform

8-260

Routines

Peter L. Walton

Submitted by: Dr. William S. Yamamoto, School of Medicine,

University of Pennsylvania, Philadelphia, Pennsylvania

TOFAST is capable of calculating in place either the discrete

Fourier transform (DFT) of real series {y_j}, j = ,0, l, ...,

N-1, where N = 2^M and M = 3,4, ... , 10, or the inverse

discrete Fourier transform (DFT^-l) of the Fourier cosine and

sine coefficients, a_k, k = ,0,l,..., n, and b_k, k = l,2,...,

n-1, respectively, where n = N/2. Output of the DFT is the

Fourier cosine and sine coefficients; output of the DFT^-l is

the real series. Several modifications are possible which

allow for input and/or output to be complex-valued.

Minimum Hardware:
8K PDP-8, (EAE optional,

depending on Floating Point

Package version)

Other Programs Needed:
Early version of DEC's FPP

Storage Requirement:
Field 0: 20-37, 64-77, 200-277,

400-1377; Field l: data array

(200-)

Restrictions:
Data array must have size 2^M,

M=3,4, ... ,10

Source Language:
MACRO-8

Catalog: July 1973

Minimum Hardware:	8K PDP-8, (EAE optional, depending on Floating Point Package version)
Other Programs Needed:	Early version of DEC's FPP
Storage Requirement:	Field 0: 20-37, 64-77, 200-277, 400-1377; Field l: data array (200-)
Restrictions:	Data array must have size 2^M, M=3,4, ... ,10
Source Language:	MACRO-8
Catalog:	July 1973