A High-precision FFT Processor Supporting 2 D FFT Algorithm

Dong YU,Li LI,Feng HAN,Kun WANG,Fan FENG,Hongbing PAN
DOI: https://doi.org/10.16592/j.cnki.1004-7859.2016.05.005
2016-01-01
Abstract:Based on the traditional DIF FFT and 2D FFT algorithm, a high-precision FFT processor supporting various input data size is designed .In the procedure of FFT calculation , a finite state machine is used as a controller .When the input data size varies in a range, the cache can be smartly managed and 1D/2D FFT algorithm is automatically chosen according to the situation whether the a-mount of input data is beyond the cache size .Therefore the whole FFT calculation can be completed without any involvement of soft -ware but a start signal .Other than the support to 2D FFT algorithm in case the cache is not enough , an optimization in the calculation procedure of twiddle factor is introduced to improve its precision and furtherly to improve the precision of final results when facing a large input data size.In the FPGA verification, a 130 dB or higher SNR(signal-noise ratio) is reached while the SNR is only around 110 dB without this optimization.
What problem does this paper attempt to address?