Quantification and reduction of Poisson-Gaussian mixed noise induced errors in ellipsometry

Area of Science:

• Optics and Photonics
• Metrology
• Statistical Analysis

Background:

• Ellipsometry is a critical metrology technique across various industries.
• Measurement accuracy in ellipsometry is compromised by Poisson-Gaussian mixed noise.
• Mueller matrix measurements are particularly susceptible to noise-induced errors.

Purpose of the Study:

• To statistically analyze and quantify errors in normalized Mueller matrix measurements caused by Poisson-Gaussian noise.
• To propose and validate a novel method for mitigating Poisson-Gaussian noise in spectroscopic ellipsometry signal demodulation.
• To experimentally characterize the noise and demonstrate the efficacy of the proposed method via simulations.

Main Methods:

• Statistical analysis to quantify error in normalized Mueller matrix measurements.
• Development of a maximum likelihood estimation (MLE) based method for noise mitigation.
• Experimental characterization of Poisson-Gaussian mixed noise using an in-house setup.
• Simulations to evaluate the performance improvement in dimension reconstruction.

Main Results:

• Quantification of measurement errors induced by Poisson-Gaussian mixed noise.
• Demonstration of a novel MLE-based method for effective noise reduction in spectroscopic ellipsometry.
• Experimental validation of noise characteristics.
• Significant improvement in dimension reconstruction accuracy shown through simulations.

Conclusions:

• The proposed maximum likelihood estimation method effectively mitigates Poisson-Gaussian noise in spectroscopic ellipsometry.
• The developed technique enhances the accuracy of normalized Mueller matrix measurements.
• This work provides a robust solution for improving the reliability of ellipsometry metrology in noisy environments.