摘要
利用遗传算法对基于半经典模型的量子细胞自动机进行仿真时,通常会遇到多个极值,容易陷入局部最优。为将量子遗传算法用于量子细胞自动机仿真,对量子遗传算法进行改进,将二进制量子位改为多进制量子位,重新设计了量子旋转门的调整策略,并给出了具体实现步骤。通过对测试函数寻优和量子细胞自动机电路的仿真,结果表明,改进后的量子遗传算法平均误差低,不易陷入局部极值,收敛速度较快,适用于量子细胞自动机仿真。
Using genetic algorithm （GA）, there are usually many extremes in the simulation of quantum cellular automatas （QCA） based on the semi-classical model and therefore it is apt to falling into local optimization. An improved quantum genetic algorithm （QGA） to displace original binary quantum bits with multistate quantum bits was proposed to use in the QCA simulation. The adjustment strategy of the quantum revolving gate was redesigned and the concrete steps to apply the improved method into the QCA simulation were also given. Through the test function optimization and QCA circuit simulation, the results show that the improved quantum genetic algorithm has many advantages, including a lower average error and a faster convergence speed, and is easy to get out local extremum. Thus the improved algorithm with more excellent performances is very suitable for the QCA simulation.
出处
《微纳电子技术》
CAS
北大核心
2011年第1期6-11,共6页
Micronanoelectronic Technology
基金
国家高技术研究发展计划项目(2008AAJ225)
