Improved Fourier and Hartley transform algorithms: application to cyclic convolution of real data
This paper highlights the possible tradeoffs between arithmetic and structural complexity when computing cyclic convolution of real data in the transform domain. Both Fourier and Hartley-based schemes are first explained in their usual form and then improved, either from the structural point of view or in the number of operations involved. Namely, we first present an algorithm for the in-place computation of the discrete Fourier transform on real data: a decimation-in-time split-radix algorithm, more compact than the previously published one. Second, we present a new fast Hartley transform algorithm with a reduced number of operations. A more regular convolution scheme based on FFT's is also proposed. Finally, we show that Hartley transforms belong to a larger class of algorithms characterized by their "generalized" convolution property.
Keywords: Algorithm design and analysis ; Telecommunication computing ; Helium ; Fourier transforms ; Fast Fourier transforms ; Discrete transforms ; Discrete Fourier transforms ; Convolution ; Arithmetic
Record created on 2005-04-18, modified on 2016-08-08