The Investigation of AutoEncoder-Based Neural Network for NMOS Circuit Anomaly Detection

被引:0
作者
Feng, Ruirong [1 ]
Wang, Zhenfan [1 ]
Hu, Yun [2 ]
Xu, Yifan [1 ]
Wang, Haohan [1 ]
机构
[1] Fujian Normal Univ, Dept Commun Engn, Fuzhou 350108, Peoples R China
[2] Fujian Normal Univ, Dept Elect Informat Engn, Fuzhou 350108, Peoples R China
来源
2024 9TH INTERNATIONAL CONFERENCE ON ELECTRONIC TECHNOLOGY AND INFORMATION SCIENCE, ICETIS 2024 | 2024年
关键词
component; NMOS; CMOS; neural network; anomaly detection;
D O I
10.1109/ICETIS61828.2024.10593795
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
This paper investigates the application of a neural network autoencoder for anomaly detection in NMOS circuits within the semiconductor industry. The focus is on the ability of the autoencoder to recognize deviations from normal operation, signifying potential anomalies. The circuits were designed using PySpice, simulating various operating conditions. Anomalies were induced by altering the load resistance and the autoencoder was trained to identify these changes through reconstruction error analysis. The model demonstrated high accuracy in anomaly detection, suggesting effective learning of the normal operational patterns. The reconstruction errors were compared to a threshold set one standard deviation above the mean to determine anomalies. Results indicated that the autoencoder could reliably identify most anomalies, providing a novel tool for circuit health monitoring and enhancing electronic device reliability. However, the model's inability to capture the full complexity of NMOS behavior under diverse conditions and the empirical nature of the threshold suggest areas for improvement. Future work will aim to increase the model's robustness, utilize a more extensive and varied dataset, and explore semi-supervised learning to detect previously unidentified anomalies.
引用
收藏
页码:216 / 220
页数:5
相关论文
共 50 条
[41]   Deep Autoencoder-Based Integrated Model for Anomaly Detection and Efficient Feature Extraction in IoT Networks [J].
Alaghbari, Khaled A. ;
Lim, Heng-Siong ;
Saad, Mohamad Hanif Md ;
Yong, Yik Seng .
IOT, 2023, 4 (03) :345-365
[43]   Graph autoencoder-based unsupervised outlier detection [J].
Du, Xusheng ;
Yu, Jiong ;
Chu, Zheng ;
Jin, Lina ;
Chen, Jiaying .
INFORMATION SCIENCES, 2022, 608 :532-550
[44]   Anomaly Detection Method for MVB Network Based on Variational Autoencoder [J].
Yang Y. ;
Wang L. ;
Chen H. ;
Wang C. .
Tiedao Xuebao/Journal of the China Railway Society, 2022, 44 (01) :71-78
[45]   Anomaly detection in random circuit patterns using autoencoder [J].
Fukuda, Hiroshi ;
Kondo, Tsuyoshi .
JOURNAL OF MICRO-NANOPATTERNING MATERIALS AND METROLOGY-JM3, 2021, 20 (04)
[46]   Anomaly detection during milk processing by autoencoder neural network based on near-infrared spectroscopy [J].
Vasafi, Pegah Sadeghi ;
Paquet-Durand, Olivier ;
Brettschneider, Kim ;
Hinrichs, Joerg ;
Hitzmann, Bernd .
JOURNAL OF FOOD ENGINEERING, 2021, 299
[47]   Deeply Integrated Autoencoder-Based Anomaly Detection and Critical Parameter Identification for Unmanned Aerial Vehicle Actuators [J].
Wang, Yan ;
Li, Shaobo ;
Yang, Lei ;
Zhang, Yizhong ;
Li, Chuanjiang ;
Zhang, Ansi ;
An, Xue .
IEEE SENSORS JOURNAL, 2024, 24 (15) :24905-24920
[48]   VASP: An autoencoder-based approach for multivariate anomaly detection and robust time series prediction with application in motorsport [J].
von Schleinitz, Julian ;
Graf, Michael ;
Trutschnig, Wolfgang ;
Schroeder, Andreas .
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE, 2021, 104
[49]   Patch Variational Autoencoder-based Industrial Defect Detection [J].
Yang, Yanqing ;
Mao, Jianxu ;
Wang, Yaonan ;
Zhang, Hui ;
Zhou, Xianen ;
Chen, Yurong .
2022 13TH ASIAN CONTROL CONFERENCE, ASCC, 2022, :674-677
[50]   A Stochastic Data Discrimination Based Autoencoder Approach For Network Anomaly Detection [J].
Aygun, R. Can ;
Yavuz, A. Gokhan .
2017 25TH SIGNAL PROCESSING AND COMMUNICATIONS APPLICATIONS CONFERENCE (SIU), 2017,