The Research And Implementation Of FFT Algorithm Based On ZSP400

Fast Fourier-transform (FFT) plays an important role in Digital Signal Processing and has been widely used in many aspects, such as the frequency spectrum analysis, the implementation of FIR digital filter, the system design and analysis, the data compaction and so on. By reason of the applicable universality, the FFT algorithm based on high-level language can be realized in PC, MCU, DSP and FPGA. Especially, the FFT algorithm based on DSP, of which the hardware characteristics and the process capability can be used, could be implemented and obtain better cost performance laying more emphasis on the methodology performance and the system resource.Taking ZSP400 for instance, this thesis discusses its architecture characteristics, instruction set characteristics and the operation mode. Using ZView tools, programed with C language and simulated in ZSP400 environment, realizing eight-point,sixteen -point DIT(decimation-in-time)and DIF(decimation-in-frequency)FFT algorithm based on radix-2 and verifying the correction of the FFT algorithm which is implemented in ZSP400. And then optimize the FFT algorithm in the program writing and algorithm simplification, introducing an advanced inverse order method and enhancing the implement efficiency of the FFT algorithm.Theoretical analysis and simulation results show that the implementation of FFT algorithm on ZSP400 core in this paper has a good performance ,and also is of certain application value.