Search

Your search keyword '"Yi Wen Hung"' showing total 4 results
Start Over Author "Yi Wen Hung" Publisher ieee
4 results on '"Yi Wen Hung"'

1. Biosignal Monitoring Clothing System for the Acquisition of ECG and Respiratory Signals

Author

Shuenn-Yuh Lee, Yi-Wen Hung, Po-Han Su, I-Pei Lee, and Ju-Yi Chen

Subjects
Biosignal acquisition, smart clothing, respiratory signal, electrocardiogram, QRS complex detection, wearable device, Electrical engineering. Electronics. Nuclear engineering, TK1-9971
Abstract

The detection of biosignals by using a comfortable material is important to improve human health. This paper presents a complete wearable system with smart clothing for the long-term monitoring of lead-I ECG and respiratory signals. The proposed system is divided into three parts, including biosignal-monitoring clothing, a biosignal acquisition device, and a software platform. The smart clothing integrates fabric-based ECG dry electrodes, conductive fiber traces, and a high-sensitivity capacitive respiration transducer to sense the biosignals. The challenge including the integration of ECG electrodes and the respiration transducer in the clothing system to provide the high-quality ECG and respiration signals for clinical use has been overcome in the proposed biosignal-monitoring clothing. The sensed signals on the smart clothing are collected in the biosignal-acquisition device through fabric-based traces and a specially designed clothing structure. Furthermore, the biosignals are processed using the biosignal-acquisition device and sent to the remote smart device through the Bluetooth module. The device according to the requirement of the front-end clothing system and the actual measurement accuracy is implemented and contributed onto the reduction in the effect of motion artifact. The software platform on the smart device provides real-time biosignal monitoring and health-information analysis. A highly efficient ECG QRS complex detection algorithm and respiratory-rate detection algorithm are also proposed. The ECG QRS complex detection algorithm is verified using the MIT/BIH Arrhythmia Database to demonstrate the achievement of high performance. The overall measured sensitivity, positive prediction, and error rate of the proposed algorithm are 99.86%, 99.93%, and 0.19%, respectively. The measured ECG signal of the clothing system is compared with the commercial silver/silver-chloride electrode by using the BIOPAC MP36 acquisition system. The result confirms the high quality of the signal, so the measured ECG signal can be used for medical applications. The function of the respiration transducer is verified by the BIOPAC SS11LA airflow transducer, and the overall accuracy of 19 test subjects is 98.74%. Using the proposed system, long-term health care in daily life can be achieved.

Published
2022
Full Text
View/download PDF

4. Accuracy Tolerant Neural Networks Under Aggressive Power Optimization

Author

Yung-Chih Chen, Xiang-Xiu Wu, Yi-Wen Hung, and Shih-Chieh Chang

Subjects
010302 applied physics, Artificial neural network, Computer science, business.industry, Deep learning, Quantization (signal processing), 02 engineering and technology, 01 natural sciences, 020202 computer hardware & architecture, Power optimization, Control theory, 0103 physical sciences, 0202 electrical engineering, electronic engineering, information engineering, Artificial intelligence, business, Critical path method
Abstract

With the success of deep learning, many neural network models have been proposed and applied to various applications. In several applications, the devices used to implement the complicated models have limited power resources, and thus aggressive optimization techniques are often applied for saving power. However, some optimization techniques, such as voltage scaling and multiple threshold voltages, may increase the probability of error occurrence due to slow signal propagation, which increases the path delay in a circuit and fails some input patterns. Although neural network models are considered to have some error tolerance, the prediction accuracy could be significantly affected, when there are a large number of errors. Thus, in this paper, we propose a scheme to mitigate the errors caused by slow signal propagation. Since the delay of multipliers dominates the critical path, we consider the patterns significantly altered by the slow signal propagation in a multiplier. We propose two methods, weight distribution and error-aware quantization to prevent the patterns from failure. Since we modify a neural network on the software side and it is unnecessary to re-design the hardware structure. The experimental results show that the proposed scheme is effective for several neural network models. It can improve the network accuracy by up to 27% under the consideration of slow signal propagation.

Published
2020
Full Text
View/download PDF

Catalog

Books, media, physical & digital resources
See catalog results