OPC Lithography Model Parameter Optimization via Random Direction Search
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for optimizing OPC lithography model parameters are inefficient in finding global optimal solutions, particularly in three and more dimensions, due to excessive computational time and difficulty in controlling meshing density, leading to suboptimal results.
Innovation Solution
A method combining a parameter configuration module, random direction method, and precise search method to quickly and precisely find optimal lens beam focus and defocus start parameters in OPC lithography models, using an objective function and constraint conditions, with the random direction method generating suboptimal combinations and the precise search method refining them into optimal solutions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If meshing is excessively dense to ensure comprehensive parameter coverage, then measurement precision is improved, but loss of time increases due to excessive computational time
Solution Approach 1:
The optimization process is segmented into two distinct stages: a random direction search stage for global exploration and a precise search stage for local optimization. This segmentation allows the system to first broadly explore the parameter space and then focus computational resources on refining the optimal solution, thereby achieving high precision without excessive computational time.
Solution Approach 2:
The patent extracts the initial parameter combination through random direction search before applying precise search methods. By separating the exploration phase from the optimization phase, the system can efficiently identify promising parameter regions and then concentrate computational effort on refining those specific regions, reducing overall optimization time while maintaining precision.
2Loss of time
If meshing is excessively sparse to reduce computational time, then loss of time is reduced, but measurement precision deteriorates due to inability to obtain global optimal solution
Solution Approach 1:
The random direction search method performs preliminary exploration of the parameter space before the precise search stage. This preliminary action identifies promising initial parameter combinations and their corresponding objective function values, providing a solid foundation for subsequent precise optimization. This ensures that even with reduced meshing density, the system can still locate the global optimal solution.
Solution Approach 2:
The patent employs feedback mechanisms where the random direction search evaluates multiple initial parameter combinations and selects those with optimal objective function values for further processing. This feedback loop ensures that the precise search stage receives high-quality initial inputs, maintaining optimization precision while reducing overall computational time.
3Device complexity
If conventional meshing methods are used for three-dimensional parameter optimization, then device complexity is reduced, but loss of time increases exponentially due to excessive number of meshes
Solution Approach 1:
The optimization method transitions from a static meshing approach to a dynamic search process. The random direction search method dynamically explores the three-dimensional parameter space by generating random initial points and evaluating their objective function values. This dynamic approach adapts to the complexity of higher-dimensional optimization problems, avoiding the exponential time increase associated with conventional meshing methods.
Solution Approach 2:
The patent changes the approach to parameter optimization by introducing random direction search that operates in the parameter space without requiring explicit meshing. This parameter change methodology allows efficient exploration of three-dimensional and higher-dimensional parameter spaces, significantly reducing optimization time while maintaining the ability to handle complex parameter relationships.
Data Source
AI summary
The present application discloses a method for optimizing OPC lithography model parameters. Preliminary suboptimal parameter combinations in proximity to a plurality of local minima of a lens beam focus BF and defocus start DS are quickly found by means of a random direction search method; and then an optimal parameter combination of the lens beam focus BF and defocus start DS is finally obtained on the basis of the suboptimal parameter combinations by means of a precise search method. An optimal parameter solution can be found quickly by combining the random direction search method and the precise search method, without artificially configuring an initial search point. Moreover, the algorithm has a high convergence rate and strong robustness, and can quickly and precisely obtain parameters such as the lens beam focus BF and defocus start DS in modeling of an OPC lithography model.