摘要
提出了一种新的整数转换为浮点数的方法。在这种方法中,求补和舍入合并实现,保留前零,舍入之后再进行规格化,减少了实现转换所需的步骤,降低了转换的复杂性。该文的设计用TSMC0.25μ标准单元库综合验证,可以工作在100MHz该文的设计已用在一个处理器的设计中。
A new method of converting integer to floating point number is proposed in this paper. In this design, implementation of complementing and rounding are combined, leading zeroes are preserved and normalization is done after rounding, so that the hardware overhead is reduced and complexity of conversion is decreased. This design is synthesized and verified with TSMC0.25um standard cell at 100MHz and is used in the design of a processor.
出处
《计算机工程》
CAS
CSCD
北大核心
2003年第19期38-39,共2页
Computer Engineering
基金
国防基金预研资助项目
关键词
浮点数
转换
舍入
Floatingpoint number
Conversion
Rounding
