摘要
共振隧穿二极管(RTD)可编程逻辑门是一种由单双稳态转换逻辑单元(MOBILE)及正、负输入分支组成的阈值逻辑电路。基于二进制神经元模型中的三层网络结构,提出了基于RTD可编程逻辑门的n变量函数实现算法。按照汉明距离由大到小的顺序,搜索最优输入向量,用定理1或定理2方法产生隐层函数,通过变换次数,确定输入向量的真假及隐层函数的权重。由于定义了最优输入向量及变换次数,提高了算法的准确性;又由于采用了定理2方法,令设计的电路更简单。
The resonant tunneling diode(RTD) programmable logic gate is a threshold logic circuit, which is composed of the monostable-bistable transition logic element(MOBILE) and positive and negative input branches. Since the three-layer network structure in the binary neuron model can be used to realize the threshold function, an realization algorithm of n-variable functions based on RTD programmable logic gate is proposed through the three-layer network structure. According to the order of the Hamming distance from large to small, the algorithm searches for the optimal input vector and uses the theorem 1 or theorem 2 to generate hidden layer functions. By judging the value of transformation times, the true and false of the input vector and the weight of the hidden layer function can be determined. The accuracy of the algorithm is improved by defining the optimal input vector and the transformation time.And due to the use of the theorem 2, the circuit designed by the algorithm is also more simple.
作者
姚茂群
冯杰
刘志强
李聪辉
YAO Maoqun;FENG Jie;LIU Zhiqiang;LI Conghui(School of Information Science and Technology,Hangzhou Normal University,Hangzhou 311121,China)
出处
《浙江大学学报（理学版）》
CAS
CSCD
北大核心
2022年第4期474-480,497,共8页
Journal of Zhejiang University（Science Edition）
基金
国家自然科学基金资助项目(61771179)。
关键词
RTD可编程逻辑门
n变量函数实现算法
汉明距离
最优输入向量
变换次数
RTD programmable logic gate
realization algorithm of n-variable functions
Hamming distance
optimal input vector
transformation time
