A uniform model for parallel fast Fourier transform (FFT) and fast discrete cosine transform (FDCT)

Extended summary form only given. The techniques for parallel processing of FFT are applied to design a parallel processing model for FDCT. Each transform is translated to a sequence of elementary operations, an equation with regularity. From the similarity of both equations, a uniform representation for FFT and FDCT is derived and a common parallel architecture (shown in diagram form) is proposed to realize both FFT and FDCT while also achieving the best known computational complexity.<>