Classification of single and double-gate nanoscale MOSFET with different dielectrics from electrical characteristics using soft computing techniques

Near-accurate classification is possible for single and double-gate nano-MOSFETs with low and high-k dielectrics based on the experimental findings of their electrical performance. Association analysis is incorporated to identify whether class determination is at all possible based on the available 28 features obtained experimentally with 800 sample data taken for four classes of MOSFETs, and FP-Growth algorithm is used to determine highest confidence rules between different subsets of classes after population analysis of data with after applying various statistical tools like t test. ReliefF algorithm is used to generate rank-wise importance of the available feature, and multi-layer perception gives best 93.33% accuracy among other classifiers. Principal Component Analysis is incorporated for creating new predictor from the existing 28 features in this work. It is found that around 95% accuracy is achieved with best 6 transformed features taken together. It is also found out that quantum capacitance per unit gate length with lowest channel diameter and highest thickness is the best feature for this classification problem. This technique may be utilized for accurate identification of higher number of classes with different transistors having different dielectrics. © 2019, Bharati Vidyapeeth's Institute of Computer Applications and Management.