Abstract:Manufacturing of semiconductor chips involves hundreds of process steps. For high-throughput semiconductor manufacturers, it is not feasible to get all the quality measurements of wafers at each step because of expensive costs. Based on the limited measurements of sampled wafers, virtual metrology (VM) allows predicting the relevant quality variables without increasing the number of physical measurements. Traditional data-driven VM methods are constructed using a predefined model along with feature extractions in data pre-processing. Often the identified VM model is not reliable if the constructed model and the extracted features are improper. In this paper, the convolutional neural network (CNN) based Gaussian process regression (GPR) VM mode (CNN-GPR) calibrated by maximization of the Bayesian posterior density distribution is systematically developed. CNN is used to extract information from massive wafer data through layers of the cascade-connected convolving filters. With the extracted features, the GPR model associates the stochastic nature of the operating steps with the quality variables to obtain the predicted means as well as the quantitated uncertainty levels. Unlike the traditional VM models where the extracted features and the prediction models are generated separately, the CNN-GPR model is sensitive to the variations of the quality variables and the features are determined simultaneously through the covariance matrices with information from both of the process and the quality variables. The proposed CNN-GPR model performs better as it represents the complex process of the whole set of measured variables. Through the studies on a real semiconductor process, the effectiveness of the proposed algorithm is demonstrated and compared against the regular regression algorithms.

Wafer-level Variation Modeling for Multi-site RF IC Testing via Hierarchical Gaussian Process

Development of Convolutional Neural Network Based Gaussian Process Regression to Construct a Novel Probabilistic Virtual Metrology in Multi-Stage Semiconductor Processes

Application of Gaussian Processes with Variable Shrinkage Method and Just-in-time Modeling in the Semiconductor Industry

Just-In-Time Modeling with Variable Shrinkage Based on Gaussian Processes for Semiconductor Manufacturing

Variation-aware gate sizing and clustering for post-silicon optimized circuits.

MMIC on-Wafer Test Method Based on Hybrid Balanced and Unbalanced RF Pad Structures

Low-cost High-Accuracy Variation Characterization for Nanoscale IC Technologies Via Novel Learning-Based Techniques.

Efficient Spatial Variation Modeling of Nanoscale Integrated Circuits Via Hidden Markov Tree

Statistical Analysis of Process Variations in RF/mm-Wave Circuits With On-The-Fly Passive Macro-Modeling

A Novel Process Variation Model for Test Cost Reduction in Wafers with Addressable Monitoring Structures

A Hierarchical Model For Characterising Spatial Wafer Variations

Semiconductor Manufacturing Process Monitoring Based on Adaptive Substatistical PCA

Comparative Study of Surrogate Modeling Methods for Signal Integrity and Microwave Circuit Applications

Bayesian Model Fusion: Enabling Test Cost Reduction of Analog/rf Circuits Via Wafer-Level Spatial Variation Modeling.

Radiated Emissions Prediction of an Electronic Packaging Based on Near-field Scanning

Compact Modeling of Process Variations in Nanosheet Complementary FET (CFET) and Circuit Performance Predictions

Efficient Spatial Variation Modeling Via Robust Dictionary Learning

Adaptive Outlier Detection for Power MOSFETs Based on Gaussian Process Regression

Reference-based Virtual Metrology method with uncertainty evaluation for Material Removal Rate prediction based on Gaussian Process Regression

A Discrete Spatial Model for Wafer Yield Prediction

Data-Efficient Prediction of Minimum Operating Voltage via Inter- and Intra-Wafer Variation Alignment