doi:

DOI: 10.3724/SP.J.1089.2010.11066

Journal of Computer-Aided Design & Computer Graphics (计算机辅助设计与图形学学报) 2010/22:9 PP.1443-1448

Reconfigurable FFT/DCT Coprocessor and Its VLSI Design


Abstract:
Fast Fourier transform-discrete cosine transform (FFT/DCT) of different lengths need to be implemented with various butterfly structures, it is therefore challenging to achieve the tradeoff between the performance and area. A new FFT/DCT structure is proposed to remedy this problem. When N is equal to 8n for N-point FFT, the area-efficient radix-2/22/23 structure is used. Otherwise the radix-2/22/23 structure is configured as a radix-8 structure, combined with performance-efficient radix-2/4 structure. Using zero-detecting and the precision adaptive mechanisms, one reconfigurable FFT/DCT coprocessor is implemented based on this structure. The coprocessor can run at 200 MHz with 148K logic gates in UMC 0.13 μm process. Experimental results show that the coprocessor can improve the FFT/DCT performance without area sacrifice.

Key words:reconfigurable,fast Fourier transform (FFT),discrete cosine transform (DCT),adaptive

ReleaseDate:2014-07-21 15:25:46



[1] Lin C T, Yu Y C, Van L D. Cost-effective triple-mode reconfigurable pipeline FFT/IFFT/2-D DCT processor[J]. IEEE Transactions on Very Large Scale Integration (VLSI) Systems, 2008, 16(8): 1058-1071

[2] Tell E, Seger O, Liu D K. A converged hardware solution for FFT, DCT and Walsh transform[C]//Proceedings of the 7th International Symposium on Signal Processing and its Applications, Paris, 2003: 609-612

[3] Sansaloni T, Pérez-Pascual A, Torres V,et al. Efficient pipeline FFT processors for WLAN MIMO-OFDM systems[J]. Electronics Letters, 2005, 41(19): 1043-1044

[4] Li X J, Lai Z S, Cui J M. A low power and small area FFT processor for OFDM demodulator[J]. IEEE Transactions on Consumer Electronics, 2007, 53(2): 274-277

[5] Ahsan M, Elahi E, Farooqi W A. Superscalar power efficient fast fourier transform FFT architecture[C]//Proceedings of the 2nd IEEE International Conference on Computer, Control & Communication, Karachi, 2009: 1-4

[6] Li M, Novo D, Bougard B,et al. Generic multiphase software pipelined partial FFT on instruction level parallel architectures[J]. IEEE Transactions on Signal Processing, 2009, 57(4): 1604-1615

[7] Franchetti F, Kral S, Lorenz J,et al. Efficient utilization of SIMD extensions[J]. Proceedings of the IEEE, 2005, 93(2): 409-425

[8] Frigo M, Johnson S G. The design and implementation of FFTW3[J]. Proceedings of the IEEE, 2005, 93(2): 216-231

[9] Markel J D. FFT pruning[J]. IEEE Transactions on Audio and Electroacoustics, 1971, 19(4): 305-311

[10] Li Zhenwei, Peng Silong, Wang Qiang. Precision reconfigurable DCT architecture and its VLSI design[J]. Journal of Computer-Aided Design & Computer Graphics, 2008, 20(3): 384-389 (in Chinese) (李振伟, 彭思龙, 王 强. 精度可配置DCT及其VLSI设计[J]. 计算机辅助设计与图形学学报, 2008, 20(3): 384-389)

[11] Cooley J W, Tukey J W. An algorithm for the machine calculation of complex Fourier series[J]. Mathematics of Computation, 1965, 19: 297-301

[12] Yuan Chen, Tsao Y C, Lin Y W,et al. An indexed-scaling pipelined FFT processor for OFDM-based WPAN applications[J]. IEEE Transactions on Circuits and Systems-II: Express Briefs, 2008, 55(2): 146-150

[13] Xiao H, Pan A, Chen Y,et al. Low-cost reconfigurable VLSI architecture for fast Fourier transform[J]. IEEE Transactions on Consumer Electronics, 2008, 54(4): 1617-1622

PDF