Buried Dielectric Modulation in Tunnel Field-Effect TransistorSu, Chonghao; Jei Liou, Juin; Yang, Shanglin
doi: 10.1088/1742-6596/2861/1/012004pmid: N/A
The tunnel field-effect transistor (TFET) with gate-on-source-only and n+ pocket configurations achieves both a low subthreshold swing (SS) and a high drain-source current. In this brief, the buried dielectric modulation effect in such a tunnel field-effect transistor (TFET) is investigated by using TCAD simulation. It is observed that the characteristic of the proposed TFET is modulated by the buried dielectric. Using a low-k dielectric material the electric field near the tunneling junction is increased, the band-to-band tunneling is enhanced and the drain-source current is increased. The simulation results show that with a low-k buried dielectric the drain-source current can be increased by more than one order of magnitude. The impact of the key parameters on the device performance is also discussed.
Remaining Useful Life Prediction of Rolling Bearings Based on Time-Frequency Enhanced Transformer ModelLin, Chunyu; Luo, Hongli; Wu, Zhenpeng; Zhang, Liangwei
doi: 10.1088/1742-6596/2861/1/012008pmid: N/A
Deep learning has achieved a lot of progress in predicting Remaining Useful Life (RUL). However, contemporary deep learning frameworks face inherent limitations, including constrained receptive fields, difficulties in capturing long-term dependencies, and singularities within the feature extraction domain. In response to these challenges, we propose a novel time-frequency enhanced Transformer model for predicting the remaining lifespan of rolling bearings. In this model, we utilize causal convolution operations to capture local features in the time-frequency domain and incorporate position encoding into the input signal. The time-frequency enhanced model exhibits end-to-end characteristics, feature fusion and selection can be performed without tedious feature extraction compared to existing methods. Furthermore, we introduce a custom Weighted Smooth Mean Squared Loss function to enhance the model’s prediction accuracy and interference resilience. In experiments, we comprehensively compare various advanced RUL prediction methods, with the results highlighting the superior performance and robust interference resistance of the time-frequency enhanced Transformer model in predicting the RUL of bearings.
Heart Failure Prediction Based on Broad Learning SystemAo, Licheng; Cai, Junyan; Lin, Yifeng; Yang, Yuer
doi: 10.1088/1742-6596/2861/1/012009pmid: N/A
Heart failure is a prevalent and serious cardiovascular condition characterized by the heart’s inability to pump blood to meet the body’s demands adequately. Current research on heart failure prediction primarily relies on conventional clinical assessment methods, traditional machine learning techniques, and traditional deep learning methods. Efficient and accurate heart failure prediction is a significant challenge due to its complex and multifactorial nature. In this study, we propose a heart failure prediction approach utilizing a broad learning system (BLS) that has the potential to capture intricate patterns in the data and enhance prediction accuracy. To evaluate our approach, we utilize an extensive dataset compiled from five previous independent datasets from the Cleveland, Hungary, Switzerland, Long Beach VA Hospital, and Stalog (heart) datasets. Experimental results demonstrate the effectiveness and efficiency of the BLS model, with a training time of 0.36 seconds and testing accuracy of 90%, precision of 88%, recall of 96%, and specificity of 82%, showcasing its potential performance for accurate heart failure prediction.
A Fetal Health Classification Approach Based on Broad Learning SystemLi, Mingxuan; Lin, Yifeng; Yang, Yuer
doi: 10.1088/1742-6596/2861/1/012010pmid: N/A
Fetal health problems are still a serious issue nowadays, affecting mothers and their children. The rate of mortality caused by fetal health problems is notably high in some areas of the earth, especially in lower-income countries. At present, besides the conventional clinical assessment, the methods of fetal health classification are mainly traditional machine learning and deep learning, such as KNN, SVM, Logistic Regression, and Naive Bayes, with the problems of low precision or long training time. In this study, we apply the broad learning system (BLS) to fetal health classification. The data set is sourced from the UCI ML repository database, which is made available by the Biomedical Engineering Institute and the University of Porto’s Faculty of Medicine. With the training accuracy of 92%, training time of 0.52s, and testing accuracy of 90%, the experimental results that highly satisfactory on BLS in the experiment.
Safe-Enhanced Autonomous Driving Technology Using Conformal Prediction ResultsLiang, Jinhao; Fang, Ruiqi; Fang, Zhenwu; Yan, Longhao; Tan, Chaopeng; Tian, Qingyun
doi: 10.1088/1742-6596/2861/1/012002pmid: N/A
Autonomous driving technology significantly improves road safety, mitigates traffic congestion, and paves the way for a more efficient and connected transportation future. However, the uncertain driving scenarios poses a great challenge to the safe control of autonomous vehicles (AVs). Therefore, this paper proposes an enhanced safety control framework by integrating the quantified prediction results of human driving behaviors into the path-planning process. First, a Long Short-Term Memory (LSTM) network is employed for driver behavior prediction, based on which, the conformal prediction, a mathematical statistical method, is introduced to quantify the uncertainties of the prediction results with a probabilistic presentation. Then, a model predictive model controller (MPC) is designed to enable the path-planning ability. By using the Lipschitz constant, the inequality for obstacle avoidance is reformulated as a hard constraint in the MPC optimization framework while considering the conformal prediction results. Finally, some simulation test cases are conducted to validate the proposed method. The results demonstrate the effectiveness to guarantee the safety of AVs with uncertain prediction results.
UAV Position Optimization for Servicing Ground Users Based on Deep Reinforcement LearningGao, Feiyu; Wang, Zichen; Liu, Xuan; Liu, Shumei
doi: 10.1088/1742-6596/2861/1/012011pmid: N/A
A Deep Q-Network (DQN) algorithm is proposed for optimization of UAVs, which can increase the communication rate of multiple users within a certain area. The UAV is able to automatically adjust its position in 3D by means of the DQN Agent’s design and training, and then get the best UAV placement location. Furthermore, the Double Deep Q-Network (Double DQN) has been researched. It has been proved that this method has higher efficiency and higher convergence rate compared with traditional DQN for locating optimal UAV position. In this paper, we perform a complex simulation of a random distribution of users in a simulated region. The experimental results indicate that UAV is able to locate the optimal position with maximum mean transfer rate, and reduce the distance from central point to customer centre, thus increasing the communication quality. This study not only verifies the effectiveness of deep reinforcement learning in dynamic positioning optimization, but also provides new insights for the optimization design of future intelligent communication systems.
Fault Detection of Rolling Bearings using Real NVP TransformationWu, Zhenpeng; Lin, Chunyu; Zhang, Rongchang; Zhang, Liangwei
doi: 10.1088/1742-6596/2861/1/012007pmid: N/A
The scarcity of rolling bearing fault data and the difficulty in annotating fault types prompt us to employ unsupervised fault detection methods. However, existing unsupervised fault detection methods rely on manual feature extraction or use indirect approaches such as reconstruction error or density approximation for fault detection. To address this issue, this paper proposes an unsupervised rolling bearing fault detection method based on a Normalizing flow model, namely the Real-Valued Non-Volume Preserving (Real NVP) model. Compared to existing methods, this approach utilizes vibration data from normal bearings for model training. It transforms the complex distribution of normal bearing data into a simple prior distribution through the Real NVP transformation. Subsequently, fault detection indicators are constructed within this simple prior distribution to achieve rolling bearing fault detection. Experiments on the benchmark bearing dataset demonstrate that the proposed method can effectively identify the fault states of rolling bearings and achieve fault detection results superior to other fault detection methods.
Prefacedoi: 10.1088/1742-6596/2861/1/011001pmid: N/A
It is our pleasure to welcome the participants and dignitaries in the International Conference on Frontiers of Electronic, Electrical and Computer Science. This conference aims to bring together leading researchers, practitioners, and industry experts to share their latest findings, discuss emerging trends, and foster collaboration in the fields of electronic, electrical, and computer science. Our conference serves as a premier platform for showcasing a diverse range of topics and methodologies, reflecting the dynamic and evolving nature of these disciplines.I would like to extend my deepest gratitude to our proceedings editor, Dr. Hu Zhu from Nanjing University of Posts and Telecommunications. His meticulous efforts in overseeing the compilation and editing of these papers have been invaluable. Dr. Zhu’s dedication and expertise have ensured that the highest standards of academic quality are maintained.We are also honored to have four distinguished keynote speakers who have graciously shared their insights and knowledge, significantly enriching our conference:Prof. Yonghui Li from the University of Sydney, for his speech: Beyond 5G towards a Super-connected WorldProf. Juin J. Liou from North Minzu University, for his speech: Broadband Impedance Measurements of EMI Filtering ChokesProf. Yang Yue from Xi’an Jiaotong University, for his speech: Machine-Learning-based Opening Object RecognitionProf. Lei Lei from Nanjing University of Aeronautics and Astronautics, for his speech: Research on Intelligent Cooperation Theory of Unmanned Aerial Vehicular Swarms Based on Digital TwinsTheir presentations have not only provided deep insights into their respective areas of expertise but also inspired new directions for future research.I would also like to acknowledge the hard work and dedication of the organizing committee, reviewers, and all the volunteers who have contributed to making this conference a success. Their collective efforts have ensured a seamless and enriching experience for all participants. Finally, I extend my appreciation to all the authors who have submitted their work to ICFEECS 2024. Your contributions are the cornerstone of this conference, and your research advances the frontiers of electronic, electrical, and computer science.We hope that the knowledge and ideas exchanged during this conference will foster collaboration and innovation, driving forward the frontiers of these exciting fields. Thank you all for your participation and support.The Committee of ICFEECS 2024List of Committee is available in this pdf.
Meta-Analysis on the Impact of Different Channel Loads on Takeover Performance in Autonomous Driving under Non-Driving TasksYao, Mingxi; Fang, Zhenwu; Zhou, Xiaozhou; Wang, Jinxiang; Yin, Guodong
doi: 10.1088/1742-6596/2861/1/012005pmid: N/A
Under conditions of high automation, drivers’ excessive engagement in non-driving related tasks can severely impact their takeover ability, posing a significant threat to road safety. Therefore, it’s essential to explore the channel resources occupied by non-driving related tasks to enhance driving safety. However, many studies have merely categorized different types of tasks in experiments without clarifying the relationship between the tasks and the channel resources they occupy. To determine whether different channel loads under non-driving related tasks have varying impacts on autonomous driving takeover performance, a meta-analysis was conducted on literature retrieved from January 2021 to January 2024. The results indicate that tasks occupying visual, auditory, motor, and cognitive channel resources during takeover all affect takeover performance. Among these, cognitive channel resource occupation significantly reduces takeover efficiency, followed by motor and auditory channels, with the visual channel having the least impact.