Fast Fourier Transform (FFT IP Core)

A Fast Fourier Transform Algorithm allows implementation of very long transforms on an FPGA using external RAM. The FFT IP Core from Mistral is designed for Run time Programmability and Optimal Resource Utilization.

Digital Down Converter, Fast Fourier Transform Algorithm, FFT IP Core, Nand Flash Controller IP Core

Overview

The Fast Fourier Transform Algorithm (FFT) is among the most important algorithms in signal processing and data analysis. Fast Fourier transform is a mathematical method for transforming a function of time into a function of frequency. The Fast Fourier Transform Algorithm computes the Discrete Fourier Transform (DFT) of a sequence, or its inverse (IDFT). The Fast Fourier Transform FPGA converts a signal from its original domain (time or space) to a representation in the frequency domain or vice versa. It is very useful for analysis of time-dependent phenomena.

Mistral’s Fast Fourier Transform Algorithm (FFT IP Core) allows implementation of very long transforms on an FPGA using external RAM. The Fast Fourier Transform Algorithm from Mistral is designed for Run time Programmability and Optimal Resource Utilization. The Fast Fourier Transform Algorithm supports run time programmable transform lengths from 256 to 1M (powers-of-2) points. The Maximum transform length is limited by the memory available.

Mistral’s FFT IP Core uses the Divide-and-Conquer approach for Fast Fourier Transform FPGA computation. This approach expresses an Fast Fourier Transform Algorithm of length N as a product of 2 integers, L x M. The L and M point FFTs are computed using a pipelined FFT block.

Higher transform lengths are supported and depend on factory configuration. The Fast Fourier Transform FPGA (FFT IP Core) from Mistral is tested on Virtex-6 FPGA from Xilinx and is designed for Run-time Programmability and Optimal Resource Utilization.

TECHNICAL SPECIFICATIONS

  • Fast Fourier Transform size, N = 2m, m = 8 to 20 (default). Can be customized
  • Forward or inverse complex transform with run time configurability
  • Run time computation of twiddle factors
  • Single precision floating point arithmetic
  • In-order input and output
  • Supports data rate of 200MS/s* (complex data)