论文部分内容阅读
针对传统频域方法用于分析金属铝互连中的噪声信号的局限性,本文采用多尺度熵方法分析电迁移噪声时间序列.结果表明在电迁移早期,噪声信号较不规律,复杂度较大;随空洞成核的发生,噪声信号规律性增强,复杂度明显减小,反映出随电迁移过程的进行系统的混乱度不断减小.通过与传统表征参量的对比,证明多尺度熵能够对电迁移失效过程进行表征.
Aiming at the limitation of traditional frequency-domain method for analyzing the noise signal in metal aluminum interconnection, the multi-scale entropy method is used to analyze the electromigration noise time series.The results show that in the early stage of electromigration, the noise signal is less regular and complex With the occurrence of void nucleation, the regularity of noise signal is enhanced and the complexity is obviously reduced, which shows that the chaos degree of the system decreases with the progress of electromigration. Compared with the traditional characterization parameters, it is proved that multiscale entropy can Electromigration failure process is characterized.