自适应阈值估计的MS模型仿真建模

Simulation-Based Multi-Signal Modeling with Adaptive Threshold Estimation

提出了一种改进的模拟电路多信号模型(MS模型)的仿真建模方法。采用Monte-Carlo方法实现电路的仿真,以获取电路特征的统计数据,利用经验公式法确定Monte-Carlo仿真的分析次数,可以减少仿真时间开销;同时,根据统计数据检验该特征的统计分布情况,再以此为依据采用自适应方法确定该特征的阈值估计区间,从而提高模型精度。最后进行了实例分析,验证了该建模方法的有效性;并从仿真时间与建模精度两个方面进行了性能比对,说明其具有的优势。

An improved simulation modeling approach of multi-signal model for analogue circuit is proposed.The Monte-carlo method is used in circuit simulation to acquire statistical data of circuit features.An empirical formula is adopted to determine the times of Monte-Carlo simulation for reducing the simulation＇s time-cost.Then,the statistical distributions of circuit features are decided by hypothesis testing using these statistical data.After that,an adaptive method is used to define the estimated threshold for features,which would improve the model＇s precision.At the end,the analysis of an example circuit is given to verify the effectiveness of this modeling method.The simulation＇s time cost and model＇s precision of different methods are compared in order to manifest the advantage of the new method.