摘要
针对以往在ISFPRM优化过程中只能处理小规模电路的不足,提出了一种新的乘积项十进制表示和处理方法来实现大电路ISFPRM面积优化.具体包括:ISFPRM多位变量的十进制数表示,基于二进制插值的极性转换方法,以及基于整数的位运算遗传算法实现ISFPRM面积优化.提出的算法能有效地避免以往算法在处理输入较多的函数时效率低下甚至无法工作的情况,算法的性能用MCNC标准电路作为测试.实验结果表明,提出的算法可以处理输入变量个数为199个的大电路,算法的速度对待处理电路的变量数不敏感特点,引入不确定项后,电路面积优化明显.
In view of the problems of the published methods of the ISPFRM functions optimization which couldn’t deal with large functions,a novel method for large ISPFRM function optimization was proposed which consists of the representation of product term in integer form,the polarity conversion method using the binary interpolation,and the circuit area optimization of ISFPRM using the bit-wise operation and the genetic algorithm.The proposed algorithm could deal with those functions with large inputs effectively,and has been implemented in C and tested under MCNC benchmarks.The experimental results show that it can deal with the large function with 199 inputs,and the speed of the algorithm is not sensitive to the functions’ polarity.After the introduction of DC terms,the circuit area is further optimized.
作者
瞿婷
王伦耀
罗文强
夏银水
QU Ting;WANG Lun-yao;LUO Wen-qiang;XIA Yin-shui(School of Information Science and Engineering,Ningbo University,Ningbo,Zhejiang315211,China)
出处
《电子学报》
EI
CAS
CSCD
北大核心
2018年第5期1101-1106,共6页
Acta Electronica Sinica
基金
国家自然科学基金(No.61471211)
国家自然科学基金重点项目(No.61131001)
关键词
不完全确定RM电路
二进制插值
位运算
GA算法
ISFPRM（Incompletely Specified Fixed Polarity Reed-Muller）
binary interpolation
bit operation
genetic algorithm
