In this paper, a new algorithm used for implementing large-point DFT is proposed. This algorithm is called large-point discrete Fourier transform (lpdft) algorithm. Firstly, when used to calculate a sequence with more...
详细信息
ISBN:
(纸本)9781538634608
In this paper, a new algorithm used for implementing large-point DFT is proposed. This algorithm is called large-point discrete Fourier transform (lpdft) algorithm. Firstly, when used to calculate a sequence with more than 100K points, the multiplication number of lpdft algorithm is reduced by 60% compared to the normal radix-2 FFT. When lpdft algorithm is implemented in FPGA, the FPGA resource occupancy rate is significantly drop due to the significant reduction of the multiplication number. Secondly, due to the use of a two-dimensional algorithm, the index value of each dimension is significantly reduced. Since the Xilinx FPGA FFT IP core function can only support calculating up to 64K points of sequence, the reduction makes it possible to use the Xilinx FPGA FFT IP core function. Simulations show that the waveforms with more than 100K points implemented with lpdft algorithm are identical in accuracy to the waveforms implemented with FFT algorithm.
暂无评论