Design of Digital Filters for Si Wafer Surface Profile Measurement - Noise Reduction by Wavelet Transform -

2010 ◽  
Vol 447-448 ◽  
pp. 544-548
Author(s):  
Masashi Ono ◽  
Kazutaka Nonomura ◽  
Li Bo Zhou ◽  
Jun Shimizu
Keyword(s):  
Wavelet Transform ◽  
Digital Filters ◽  
Semiconductor Industry ◽  
Surface Profile ◽  
Measured Data ◽  
International Standards ◽  
Wafer Surface ◽  
Profile Measurement ◽  
Severe Problem ◽  
Low Pass

Recently in semiconductor industry, production of ever flatter, thinner and larger silicon wafers are required to fulfill the demands of high-density packaging and cost reduction. In geometric evaluation of Si wafers, according to SEMI (Semiconductor Equipment and Materials International) standards, the required wafer flatness approaches to the 22 nanometers by year 2016 [1]. For such application, uncertainty of measured data is encountered as a severe problem because high resolution instrument always incorporate a certain degree of noise. In order to precisely evaluate the wafer profile, it is essential to remove the noise from the measured data. Described in this paper is design and development of digital filters for denoising. Compared to the conventional low-pass filters, the developed filter by use of wavelet transform not only provides better performance of decomposition in the spatial frequency domain, but also offers the new capability of denoising in amplitude domain.

Design of Digital Filters for Si Wafer Surface Profile Measurement – Noise Reduction by Lifting Scheme Wavelet Transform

2010 ◽  
Vol 126-128 ◽  
pp. 732-737 ◽  
Author(s):  
Kazutaka Nonomura ◽  
Masashi Ono ◽  
Li Bo Zhou ◽  
Jun Shimizu ◽  
Hirotaka Ojima
Keyword(s):  
Wavelet Transform ◽  
Digital Filters ◽  
Semiconductor Industry ◽  
Surface Profile ◽  
Lifting Scheme ◽  
Measured Data ◽  
International Standards ◽  
Wafer Surface ◽  
Profile Measurement ◽  
Severe Problem

Recently in semiconductor industry, production of ever flatter, thinner and larger silicon wafers are required to fulfill the demands of high-density packaging and cost reduction. In geometric evaluation of Si wafers, according to SEMI (Semiconductor Equipment and Materials International) standards, the required wafer flatness approaches to the 22 nanometers by year 2016 [1]. For such application, uncertainty of measured data is encountered as a severe problem because the requirement has met the limit of available instrument in terms of resolution and reliability. In order to precisely evaluate the wafer profile, it is essential to remove the noise from the measured data. Described in this paper is design and development of digital filters for denoising. In previous paper, digital filters for denoising with Haar wavelet transform are described. In this paper, the new filters by use of 2nd generation wavelet transform (lifting scheme) are proposed and show better performance of decomposition in the spatial frequency domain and amplitude domain.

Research on Digital Filters for Si Wafer Surface Profile Measurement - Design of Filters by Total Variation

2012 ◽  
Vol 565 ◽  
pp. 656-661
Author(s):  
Hirotaka Ojima ◽  
Kazutaka Nonomura ◽  
Li Bo Zhou ◽  
Jun Shimizu ◽  
Teppei Onuki
Keyword(s):  
Total Variation ◽  
Digital Filter ◽  
Digital Filters ◽  
Large Population ◽  
Surface Profile ◽  
Wafer Surface ◽  
Profile Measurement ◽  
Actual Measurement ◽  
True Value ◽  
New Tv

The underlying data form of a wafer is a matrix of length (or height) measurements. In the presence of noise, evaluation parameters are normally biased. The expectation value such as peak-to-valley and GBIR (global backside ideal range) is systematically larger than the “true” value. Correction and compensation need a large population of measurements to analytically estimate both bias and the uncertainty. In this study, approach to obtain the true value is to extract a “true” profile by filtering noise from the measured data. In previous paper, the digital filter with wavelet transformation (WT) is proposed and efficiency to remove the noise, however, the method is introduced the pseudo-Gibbs effect. Then, we propose the digital filter with new algorithm of total variation (TV). In this paper, the new algorithm of TV is proposed and the digital filter by new TV indicate that data is filtered without the pseudo-Gibbs effect. The digital filters by WT and new TV are applied on the sample data of actual measurement system to investigate their performance of noise reduction.

Intelligent Profile Measurement for Wide-Area Resist Surface Using Multi-Sensor AFM System

2008 ◽  
Vol 381-382 ◽  
pp. 407-410
Author(s):  
Shu Jie Liu ◽  
K. Watanabe ◽  
Satoru Takahashi ◽  
Kiyoshi Takamasu
Keyword(s):  
Thin Film ◽  
Semiconductor Industry ◽  
Surface Profile ◽  
Vertical Direction ◽  
Profile Measurement ◽  
Measuring Device ◽  
Scanning Method ◽  
Angle Measuring ◽  
Experimental Apparatus ◽  
Sensor Errors

In the semiconductor industry, a device that can measure the surface-profile of photoresist is needed. Since the photoresist surface is very smooth and deformable, the device is required to measure vertical direction with nanometer resolution and not to damage it at the measurement. We developed the apparatus using multi-cantilever and white light interferometer to measure the surface-profile of thin film. But, this system with scanning method suffers from the presence of moving stage and systematic sensor errors. So, in this paper, an error separation approach used coupled distance sensors, together with an autocollimator as an additional angle measuring device, was consulted the potentiality for self-calibration of multi-cantilever. Then, according to this method, we constructed the experimental apparatus and do the measurement on the resist film. The results demonstrated the feasibility that the constructed multi-ball-cantilever AFM system combined with an autocollimator could measure the thin film with high accuracy.

Design of Digital Filters for Si Wafer Surface Profile Measurement - Denoising by Total Variation

2012 ◽  
Vol 516 ◽  
pp. 332-336
Author(s):  
Hirotaka Ojima ◽  
Kazutaka Nonomura ◽  
Li Bo Zhou ◽  
Jun Shimizu ◽  
Teppei Onuki
Keyword(s):  
High Resolution ◽  
Spatial Frequency ◽  
Frequency Domain ◽  
Total Variation ◽  
Digital Filters ◽  
Measured Data ◽  
Design Rule ◽  
Profile Measurement ◽  
Amplitude Component ◽  
Spatial Frequency Domain

In the semiconductor industry, high resolution and high accuracy measurement is needed for the geometric evaluation of Si wafers. The flatness parameters are important to evaluate the wafer profile and are required to be the same level as the design rule of IC, and the tolerance for flatness is very tight. According to SEMI (Semiconductor Equipment and Materials International) standards, the required wafer flatness will be 22 nanometres by the year 2016. However, to obtain a higher resolution for sensors, the uncertainty becomes very large compared to the resolution and influences the measured data when the noise is increased. High resolution instruments always incorporate a certain degree of noise. In the presence of noise, form parameters are normally biased. Correction and compensation need a large population of measurements to analytically estimate both bias and uncertainty. The estimation is still far from perfect because of the nature of noise. Another approach is to extract a true profile by filtering noise from the measured data. For the purpose of noise reduction, low-pass filters by Gaussian smoothing and Fourier transform are often used. The noise is normally considered to be a component of small deviation (amplitude) with high frequency which also takes a normal distribution around zero. However these conventional filters can remove the noise in the spatial frequency domain only. So, it is essential to design a filter capable of removing the noise both in the spatial frequency domain and the amplitude component. Thus, we have designed and developed new type of digital filter for denoising. We introduce two new digital filters. One is wavelet transform capable of denoising in the spatial frequency domain and amplitude component, and the other is total variation that can be applied to discontinuous signals without introducing artificial Gibbs Effects.

scholarly journals Low-Pass Filtering Empirical Wavelet Transform Machine Learning Based Fault Diagnosis for Combined Fault of Wind Turbines

Entropy ◽  
2021 ◽  
Vol 23 (8) ◽  
pp. 975
Author(s):  
Yancai Xiao ◽  
Jinyu Xue ◽  
Mengdi Li ◽  
Wei Yang
Keyword(s):  
Machine Learning ◽  
Wavelet Transform ◽  
Fault Diagnosis ◽  
Wind Turbine ◽  
Wind Turbines ◽  
Gear Tooth ◽  
Support Vector ◽  
Grey Wolf Optimizer ◽  
Empirical Wavelet Transform ◽  
Low Pass

Fault diagnosis of wind turbines is of great importance to reduce operating and maintenance costs of wind farms. At present, most wind turbine fault diagnosis methods are focused on single faults, and the methods for combined faults usually depend on inefficient manual analysis. Filling the gap, this paper proposes a low-pass filtering empirical wavelet transform (LPFEWT) machine learning based fault diagnosis method for combined fault of wind turbines, which can identify the fault type of wind turbines simply and efficiently without human experience and with low computation costs. In this method, low-pass filtering empirical wavelet transform is proposed to extract fault features from vibration signals, LPFEWT energies are selected to be the inputs of the fault diagnosis model, a grey wolf optimizer hyperparameter tuned support vector machine (SVM) is employed for fault diagnosis. The method is verified on a wind turbine test rig that can simulate shaft misalignment and broken gear tooth faulty conditions. Compared with other models, the proposed model has superiority for this classification problem.

Sign in / Sign up

Export Citation Format

Share Document