A note on the computational complexity of the arithmetic Fourier transform
Nazif Tepedelenlioglu · IEEE Transactions on Acoustics Speech and Signal Processing · 1989
It is shown that the number of data points the arithmetic Fourier transform (AFT) needs for an N-point Fourier transform is proportional to N/sup 2/. Thus, for example, while a standard fast Fourier transform algorithm requires 1024 samples to yield 1024 spectral components, AFT would take more than 300000 samples to do the same job.>