一种FFT并行处理机的设计与实现

doi:10.3969/j.issn.1001-2400.2010.04.009

Abstract

Abstract:

The ASIP combines programmability of DSP and high speed of ASIC, and the parallel array processing system based on it plays an important role in the high-speed and real-time processing applications. An implementation of 1024 points FFT parallel processor based on the ASIP is presented in this paper. An ASIP with the reduced instruction set computer (RISC) architecture is designed as the processing element (PE), which constructs the kernel of the parallel processing system. In addition, the communication matrix is adopted to achieve the data exchange between PEs. Experimental results show that, compared with the typical DSP, the new proposed architecture improves the processing speed by 30％. Furthermore, this system has the advantages of larger parallel scale, greater function adaptability, relatively lower designing complexity and greater design reuse ability. So it can be achieved in a single FPGA in the form of SoC.

Key words: application specific instruction set processor(ASIP), fast Fourier transform(FFT), reduced instruction set computer (RISC), parallel processing, data communication

ZHANG Li;LI Shuang-fei;SHI Guang-ming;LI Fu. Design and implementation of the FFT parallel processor based on ASIP[J].J4, 2010, 37(4): 630-635.

References

[1] Kreutzer K, Malik S. From ASIC to ASIP: the Next Design Discontinuity [C]//IEEE International Conference on Computer Design: VLSI in Computers and Processors. Washington: IEEE, 2002: 84-90.
[2] 刘红侠, 杨靓, 黄巾, 等. 可变长FFT并行旋转因子高效产生算法及实现[J]. 西安电子科技大学学报, 2009, 36(3): 541-546.
Liu Hongxia, Yang Liang, Huang Jin, et al. Effective Algotithm of Parallel Twiddle Factor Generation for Programmable FFT Processing and Its Implemetation [J]. Journal of Xidian University, 2009, 36(3): 541-546.
[3] Jiang R M. An Area-efficient FFT Architecture for OFDM Digital Video Broadcasting [J]. IEEE Trans on Consumer Electronic, 2007, 53(4): 1322-1326.
[4] Shin M, Lee H. A High-speed Four-parallel Radix-24 FFT/IFFT Processor for UWB Applications [C]//IEEE International Symposium on Circuits and Systems. Washington: IEEE, 2008: 960-963.
[5] 王旭东. 固定几何结构的FFT算法及其FPGA实现[C]//2003嵌入式世界研讨暨展示会. 北京: 中国计算机学会, 2003: 274-276.
[6] 姬红兵, 谢维信, 贺伟. 一种基于数字信号处理器的有效FFT实现[J]. 西安电子科技大学学报, 1998, 25(4): 445-449.
Ji Hongbing, Xie Weixin, He Wei. A Effective Implementation of FFT Based on DSP[J]. Journal of Xidian University, 1998, 25(4): 445-449.
[7] Jain M K, Balakrishnan M, Kumar A. ASIP Design Methodologies: Survey and Issues [C]//The 14th International Conference on VLSI Design. Bangalore: IEEE/ACM, 2001: 76-81.
[8] Li Z, Shuangfei L. A Research on an ASIP Processing Element Architecture Suitable for FPGA Implementation [C]//IEEE International Conference on Computer Science and Software Engineering: Vol 3. Wuhan: IEEE, 2008: 441-445.
[9] 刘书明, 苏涛, 罗军辉. TigerSHARC DSP应用系统设计[M]. 西安: 西安电子科技大学出版社, 2003.

Design and implementation of the FFT parallel processor based on ASIP

PDF (PC)

Like

Knowledge

Cited

Abstract

Cite this article

share this article

References

Related Articles 5

Metrics

Comments

Recommended 10

[1]	ZHAO Baowei;XIANGLI Bin;Lü Qunbo;ZHANG Guifeng;ZENG Xiaoru;GUO Wenji. Parallel RX algorithm implementation based on the FPGA and multi-DSP system [J]. J4, 2014, 41(3): 152-156.
[2]	WANG Xing;GONG Shuxi;WANG Wentao;ZHANG Yujie. Fast analysis of electromagnetic properties of surface/surface configurations [J]. J4, 2011, 38(4): 106-111.
[3]	LIU Hong-xia;YANG Liang;HUANG Jin;HUANG Shi-tan. Effective algorithm of parallel twiddle factor generation for programmable FFT processing and its implementation [J]. J4, 2009, 36(3): 541-546.
[4]	SU Tao;ZHUANG De-jing;WU Shun-jun. Fast algorithm for real time SAR imaging on a parallel processor [J]. J4, 2005, 32(1): 21-25.
[5]	ZHAO Bo;DU Jian-chao;YAN Yao-ping. A block-matching algorithm based on hardware implementation and its VLSI architecture [J]. J4, 2003, 30(2): 160-164.