Robust bandgap circuit design enabled with a fast and accurate variation-aware verification
STMicroelectronics uses Solido Variation Designer to address verification challenges
Solido Variation Designer software enabled the STMicroelectronics Smart Power design team to debug and perform a thorough and accurate verification of the bandgap voltage reference circuit and achieve a yield greater than eight sigma. The team arrived at these results with only 8,800 SPICE accurate simulation runs, equivalent to greater than 10 billion brute force Monte Carlo simulations for the same accuracy.
Related resources
Machine learning-powered etch bias prediction for etch retargeting flow enhancement
The paper proposes a machine learning (ML) based approach to predict etch bias, which can replace the traditional rule-based etch bias tables
Guided random synthetic layout generation and machine-learning based defect prediction for leading edge technology node development
Combining synthetic layout generation and machine learning accelerates advanced semiconductor development. It enables early identification of process hotspots and defect prediction, improving efficiency and reducing costs.
A study on the improvement of machine learning (ML)-based defect prediction model reflecting process variations
This paper presents an ML-based risk pattern predictor that combines pattern segmentation, Greedy sampling, and unbiased statistics to find defects and improve yield.