串扰目标识别框架  

Crosstalk Target Identification Framework

作　　者：张富彬[1] 李长青[1] Ching-Yen Ho 彭思龙[1] 

机构地区：[1]中国科学院自动化研究所国家专用集成电路设计工程研究中心 [2]Synopsys Inc.Mountain View CA 94043

出　　处：《计算机工程与应用》2006年第30期72-77,共6页Computer Engineering and Applications

摘　　要：提出了一个全面识别静态和动态串扰噪声的算法框架,在静态串扰噪声的识别中,通过静态串扰噪声的峰值以及噪声宽度信息来识别串扰目标。在动态串扰噪声的识别中,引入了混合时序分析,缩小了时间窗区间,并通过测试生成来验证信号间的逻辑关系,通过这些准确的时序及逻辑信息,识别出串扰目标。A framework is presented which consists a set of algorithms to identify static and dynamic crosstalk noise.In the course of static crosstalk noise identification,the static crosstalk noise peak and noise width are used to identify crosstalk target.When identifying dynamic crosstalk noise,bybrid timing analysis is used to provide accurate signal arrival time,at the same time,a novel test generation is chosen to verify the correlation of signals.So dynamic crosstalk target can be identified by these accurate timing and logic correlation information.

关 键 词：静态串扰噪声 动态串扰噪声 串扰识别 混合时序分析 测试生成 

分 类 号：TP391.72[自动化与计算机技术—计算机应用技术]