Designed for fast fourier transformation. Fourier-domain convolution of 2D-data-sets, the PowerFFT processor computes a 1k complex FFT within 10µs. By implementing multiple processors and additional DRAM into a multiprocessor-system, almost linear speed-up of system-performance can be achieved.
Targeting at the transformation of real-valued data, the processors functional scope is extended by dedicated real-valued FFT algorithms, allowing for simultaneous transformation of 2 real-valued sequences with one complex FFT. With this, concerning real-valued data, the performance of the FFT processor is further improved.
The benchmarks for the FFT processor with double-precision floating point computation are:
- Benchmarks 1k pts points FFT, including windowing 10 µs
- 1M points convolution or correlation, including windowing 21 ms
- 1k points convolution or correlation, fixed filter 23 µs
- 1k points convolution or correlation, updated filter 33 µs
- 1M points convolution or correlation, fixed filter 42 ms
- 1M points convolution or correlation, updated filter 63 ms
- 1K x 1K points 2D FFT, including windowing 21 ms
- 1K x 1K pts 2D convolution or correlation, fixed filter 42 ms
- 1K x 1K pts 2D convolution or correlation, updated filter 63 ms
This combination of fast FFT-computation and high precision opens a vast field of applications to the FFT processor.
If you like to receive specific information or whitepapers then please mark the checkboxes and fill the form fields.
