Computational lithography method for model-driven convolution neural network

Abstract

The invention discloses a computational lithography method for a model-driven convolution neural network (MCNN), and the method can improve the computation speed and convergence performance of OPC (optical proximity correction) method. The technical scheme includes: expanding and truncating the gradient iterative algorithm, and constructing a model-driven convolution neural network (MCNN); based on an imaging model of a lithography system, constructing a decoder corresponding to the MCNN; bringing the MCNN and the decoder to end-to-end connection, and subjecting the MCNN to the following training: optimizing the parameters of the MCNN by back propagation algorithm to minimize the error between the input data of MCNN and the decoder; separating the decoder from the MCNN at the end of the training; inputting a to-be-optimized circuit layout into the trained MCNN so as to obtain an estimated result of an OPC mask; taking the estimated result of OPC mask as the initial value, and carryingout iterative updating on the set number of times of the mask by gradient iterative algorithm, thus obtaining the final OPC mask optimization result.

Description

technical field [0001] The invention relates to the technical field of computational imaging, in particular to a computational lithography method based on a model-driven convolution neural network (MCNN). Background technique [0002] Lithography is one of the core technologies used to manufacture VLSI. The lithography system uses a light source to illuminate the mask, and the integrated circuit layout on the mask is reproduced on the silicon wafer through the projection objective lens. At present, the semiconductor industry mainly uses computational lithography to improve the resolution and imaging quality of lithography systems. Optical proximity correction (OPC for short) is one of the important computational lithography techniques. OPC technology modulates the amplitude of the light wave transmitted through the mask by modifying the mask pattern or adding necessary auxiliary patterns on the mask pattern, thereby compensating for imaging distortion caused by diffraction...

AI Technical Summary

Problems solved by technology

[0003] The existing gradient-based OPC optimization algorithm optimizes the transmittance of each pixel of the mask through cyclic iterations. In order to obtain the optimization result of a mask, a large number of iterations are often required. Calculate the lithographic imaging error corresponding to the current mask pattern, so the calculation complexity of the algorithm is high, and the operation takes a long time

In addition, due to the nonlinear characteristics of the imaging model of the lithography system, the existing gradient algorithm is easy to fall into the local optimal solution of the OPC optimization problem, resulting in large convergence errors and affecting the imaging quality of the lithography system.

[0004] In summary, the existing gradient-based OPC methods need to be further improved and improved in terms of calculation speed and convergence performance.

Images