Method of Computing the DFT of N-Point Real Data Sequence
Original Publication Date: 1980-Jul-01
Included in the Prior Art Database: 2005-Feb-13
In many practical applications, one has to compute the Discrete Fourier Transform (DFT) of an N-point real data sequence. The conventional approach to this computation consists in evaluating simultaneously the transforms of two consecutive real data sequences with a single complex DFT of dimension N . While this approach is reasonably efficient from the standpoint of number of operations, it has the disadvantage of introducing a one-block delay in the computation of the DFT. In this article, we introduce an efficient method which allows the computation of the transform of a real N-point sequence with a single complex DFT of dimension N/2. We show that this method is computationally efficient and does not introduce extra delays in the evaluation of the DFT.