PCA - Wavelet Coefficients for T2 chart to Detect Endpoint in CMP process

The development of the semiconductor industry, with advances in sensors oblige us to deal with largedatasets do not stop increasing, while monitoring devices are becoming more and more complexes andsophisticated. As the measurement points become closer. Among the complex monitoring process, thedetection of the end of polishing (EPD) during the chemical mechanical planarization (CMP) process isconsidered as a critical task in semiconductor manufacturing. In this paper, a sequence of an Acousticalemission (AE) waveform signals are collected during the progression of the CMP process will bemonitored using PCA- Wavelet daubechie Coefficients based on T2chart. In order to detect theendpoint, we should not only to remove the noise from the obtained acoustical waveform signal, butalso reduce dimensionality of monitored coefficient, by employing discrete wavelet algorithmfor cleaning signals , Principal component analysis (PCA) for reducing dimensionality . Also, acomparative study is presented to show the out-performance of PCA-Wavelet Hotelling chart in detectingthe endpoint.