An unified algorithm for sinusoid-class orthogonal transforms
Baoyu Zheng · 2002
The DFT is an important tool in digital signal processing. An effective index mapping is introduce by which the odd-length sinusoid-class orthogonal transforms, such as the generalised discrete Fourier transforms (GDFTs), generalised discrete Hartley transforms (GDHTs), discrete cosine transforms (DCTs) and discrete sine transforms (DSTs), can be converted to DFTs, real-valued DFTs (RDFTs) or DHTs of the same length, using primarily permutations and sign changes only. The algorithm proposed is more efficient than others previously published.