发明名称 PRINTING PROCESS CALIBRATION AND CORRECTION
摘要 Various embodiments include approaches for calibrating a model for a lithographic printing process. Some embodiments include a computer-implemented method for calibrating a model for a lithographic printing process. Some approaches include: identifying parameters for a model of the lithographic printing process; assembling a population of design content including potentially printable features that can be printed by the lithographic printing process; preparing at least one matrix expressing a similarity between the potentially printable features in terms of the parameters for the model; determining a manifold of smaller dimensionality than the parameters for the model which exhibit maximum variation in similarity within the at least one matrix; and selecting a sample dataset of the potentially printable features from the manifold.
申请公布号 US2014380255(A1) 申请公布日期 2014.12.25
申请号 US201313921378 申请日期 2013.06.19
申请人 International Business Machines Corporation 发明人 Barai Samit;Rosenbluth Alan E.
分类号 G06F17/50 主分类号 G06F17/50
代理机构 代理人
主权项 1. A computer-implemented method of sampling lithographic content producible by a lithographic printing process, the method performed using at least one computing device, the method comprising: identifying parameters for a model of the lithographic printing process; assembling a population of design content including potentially printable features that can be printed by the lithographic printing process; preparing at least one matrix expressing a similarity between the potentially printable features in terms of the parameters for the model, wherein the at least one matrix is formed from similarity weights that express similarity relationships between neighboring features; determining a manifold of smaller dimensionality than the parameters for the model which exhibit maximum variation in similarity within the at least one matrix, using the at least one computing device, wherein the manifold of smaller dimensionality is determined from low-dimensioned coordinate assignments for the potentially printable features that reproduce similarity relationships between neighboring features within a low-dimensioned space; and selecting a sample dataset of the potentially printable features from the manifold.
地址 Armonk NY US