The Fast Fourier Transform (FFT) is a mathematical method widely used in signal processing in sonic,
spectral and imaging analysis applications. It is applied in a wide variety of applications including
biomedical engineering, mechanical analysis, analysis of stock market data, geophysical analysis, and
conventional radar communications field. An FFT is an optimized version of a DFT (discrete Fourier Transform)
used to transform data from the time-domain to the frequency-domain.
This FFT Accelerator runs on DRC's AcceliumTM Coprocessor and is made available to software applications through a standard API call mechanism. The current version 1D FFT can perform FFTs of lengths: 4, 16, 64, 256 and 1024. Future versions are planned that will extend the length and support to 2D implementations.
The DRC FFT Accelerator enables immediate use of the per-coded and proven implementation versus taking 6 months to develop the routine. The cost savings and time to market advantages to your organization can be substantial.
The 1K 1D FFT Accelerator is based on a Radix-4, Decimation-In-Time algorithm.
It consists of six major components:
Partial Results Buffer