Analog Verification and Characterization with Monte Carlo and High-Sigma Analysis
Analog IP developers, managers, CAD managers
Semiconductor companies designing ICs for smart phones, automotive and industrial applications, CPUs, GPUs and memory components all employ teams of custom IC designers to create the highest performance chips that are as small as possible, and at the lowest costs. Designers must verify and characterize their IPs' sensitivity to random parametric variations in the manufacturing process (both on-chip variation and die-to-die variation), as it can have a large impact on a circuit's performance or even cause functional failures. Accuracy and efficiency of statistical analysis techniques is a main topic of interest here because many methods may require a large simulation effort.
In this webinar we will show
* advanced Monte Carlo sampling methods
* high-sigma analysis methods using machine learning (modeling) and optimization methods, with high-sigma analysis for large circuits with many 10,000s of devices
* statistical verification of full custom circuits for functional defects