基于支持向量机及遗传算法的光刻热点检测

摘要

A lithography hotspot detection method based on support vector machine (SVM) and genetic algorithm (GA) is proposed. Frequency domain features of integrated circuit (IC) layout samples are first extracted with discrete cosine transformation. Then SVM classifiers are trained with the layout samples so that it can be used for hotspot detection. To improve the precision and efficiency of this method, genetic algorithm is used for feature selection and SVM parameter optimization. Experiment results show that the proposed method can improve the precision of lithography hotspot detection effectively.%提出一种基于支持向量机(SVM)及遗传算法(GA)的集成电路版图光刻热点检测方法.首先对版图样本进行离散余弦变换(DCT)以提取样本的频域特征,然后基于这些样本训练SVM分类器以实现对光刻热点的检测.为了提高光刻热点检测的精度及效率,采用遗传算法(GA)对频域特征进行选择,并同时优化SVM参数.实验结果表明,基于SVM及版图频域特征并结合遗传算法进行优化的光刻热点检测方法可以有效提高版图光刻热点的检测精度.