Pipeline Design of Transformation between Floating Point Numbers Based on IEEE754 Standard and 32-bit Integer Numbers
暂无分享,去创建一个
A pipeline design method for transformation between floating point numbers based on IEEE754 standard and 32-bit integer numbers was proposed. There are two kinds of pipelines. One is for double-precision or single-precision floating point numbers transform into integer numbers and other is for integer numbers transform into double precision or single-precision floating point numbers. In this method hardware cost was reduced through double-precision floating point numbers and single-precision floating point numbers sharing one data path. Shifting and rounding was simplified through setting the initial value of exponent. Special numbers were also discussed and corresponding signals were defined in this method. The leading zero detection method for this method was described and a leading zero detection unit for 32-bit numbers was designed. A module of the pipeline design method proposed in this paper was designed with VHDL and EP2C35F484C6 is the target device of implementation. The source utilization and maximum frequency of this module were estimated. The simulation results were discussed. The quality of this method was confirmed by these results.
[1] Ansi Ieee,et al. IEEE Standard for Binary Floating Point Arithmetic , 1985 .
[2] Javier D. Bruguera,et al. Leading-one prediction scheme for latency improvement in single datapath floating-point adders , 1998, Proceedings International Conference on Computer Design. VLSI in Computers and Processors (Cat. No.98CB36273).
[3] Kai Hwang,et al. Computer arithmetic: Principles, architecture, and design , 1979 .
[4] Guido D. Salvucci,et al. Ieee standard for binary floating-point arithmetic , 1985 .