Search

Layout hotspot detection is one of the most important issues for the reliability enhancement of integrated circuits. Machine learning-based hotspot detectors have shown their advantages of efficiency and generalization compared with computationally intensive lithography process simulation. However, most machine learning-based hotspot detectors only accept layout clips of fixed size as input with the potential defect whose location is restricted at the center of each clip. Therefore, they cannot be used directly for multiple hotspots detection in a large area, which occurs frequently in real design cases. In this article, we build a new end-to-end hotspot detector based on a fully convolutional network, which has the flexibility of detecting a various number of hotspots in a layout of any size at one time. Moreover, we also develop a transfer learning scheme matching our proposed detector network, which can reduce the requirement of sample number when setting up a new model for a more advanced technology node. The experimental results demonstrate our proposed hotspot detector outstanding among state-of-the-art works and the transfer learning scheme is effective. [ABSTRACT FROM AUTHOR]

As an approach of resolution enhancement technique (RET), sub-resolution assist feature (SRAF) is used to meet the pitch requirements for sparse patterns in off-axis illumination (OAI) and supplements the spatial frequencies of main patterns to improve the resolution. Rules-based SRAF (RB-SRAF) is a technique that foundry extracts SRAF placement rules from wafer data or simulation data and then apply these rules to full chip. RB-SRAF often produces a lot of conflicts. Usually, the geometric constrains are the only factors taken into account in cleaning up of these conflicts, which may lead to mistaken cleanup and be counter-productive. Therefore, it is necessary to introduce some physical factors in the course of cleanup. The transmissivity distribution of mask in hotspot region can be obtained by inverse lithography technology (ILT). And the statistical information of the grayscale map from ILT can be used for leading cleanup and keeping the SRAFs which contribute more to process window (PW). The feasibility of this method has been verified by a series of simulations. [ABSTRACT FROM AUTHOR]