Addressing technique for parallel memory accessing in radix-2 FFT processors

Abstract

This paper presents an efficient technique for addressing in radix-2 FFT architectures. The novel addressing organization provides parallel load and store of the data involved in a radix-2 butterfly computation. The addressing scheme is based on a permutation of the FFT data, which leads to the minimization of the address generating circuit and the butterfly processor control. The paper proves the correctness of the technique and includes a FPGA implementation.