高速浮点运算单元的FPGA实现

被引:8
作者
张小妍
邵杰
机构
[1] 南京航空航天大学信息科学与技术学院
关键词
浮点运算; FPGA; 流水线技术;
D O I
暂无
中图分类号
TN791 [];
学科分类号
摘要
运用流水线技术对单精度浮点乘法和加法运算单元进行了优化设计。浮点加法器采用了改进的双路径结构,重点对移位单元和前导1检测单元的结构进行了优化。浮点乘法器在对被乘数进行Booth编码后,采用改进的4-2压缩器构成Wallace树,在简化逻辑的同时,提高了系统的吞吐率。经过仿真验证,在Virtex-4系列FPGA(现场可编程门阵列)上,浮点加法器的最高运行速率达到405 MHz,浮点乘法器的最高运行速率达到429 MHz。
引用
收藏
页码:24 / 27+30 +30
页数:5
相关论文
共 8 条
  • [1] Analysis of high-per-formance floating-point arithmetic on FPGAs. GOVINDUG,ZHUO L,CHOI S,et al. Proceed-ings of the 18th International Parallel and Distributed Process-ing Symposium . 2004
  • [2] Design of a field-programmable du-al-precision floating-point arithmetic unit. DINIZ P C,GOVINDUG. Proceedings of16th International Conference on Field Programmable Logicand Applications . 2006
  • [3] Open source high per-formance floating-point modules. HEMMERTK S,UNDERWOOD K D. Proceedings of the 14thAnnual IEEE Symposium on Field-Programmable CustomComputing Machines . 2006
  • [4] An IEEE compliantfloating-point adder with the deeply pipelining paradigm onFPGAs. SHAO Jie,YE Ning,ZHANG Xiaoyan. Proceedings of 2008 International Conference onComputer Science and Software Engineering . 2008
  • [5] Effective implementation of float-ing-point adder using pipelined LOP in FPGAS. MALIK A,KO Seok-Bum. Proceed-ings of Canadian Coufereuce on Electrical and Computer Engi-neering . 2005
  • [6] Novel and efficient 4∶2 and 5∶2 compressors with minimumnumber of transistors designed for low-power operations. GOPINEEDI P D,THAPLIYAL H,SRINIVAS M B,et al. Proceedings of the 2006 International Conference on Embed-ded Systems and Applications . 2006
  • [7] Leading-one prediction with con-current position correction. Bruguera,J,Lang,T. IEEE Transactions on Computers . 1999
  • [8] Design of verydeep pipelined multipliers for FPGAs. PANATO A,SILVA S,WAGNER F,et al. Proceedings ofDesign,Automation and Test in Europe Conference and Exhi-bition . 2004