检索结果-内蒙古大学图书馆

Distributed data-privacy preserving federated learning method for sleep stage classification

Biomedical Signal Processing and Control 2025年 109卷

作者： Li, Hongjin Zhao, Xiaohui Liu, Hang Wang, Ziyi Cai, Yukai Yang, Chuanshuai Cong, Fengyu School of Biomedical Engineering Faculty of Medicine Dalian University of Technology Dalian China Liaoning Key Laboratory of Integrated Circuit and Biomedical Electronic System Dalian University of Technology Dalian China Department of Respiratory and Critical Care Medicine The Affiliated Central Hospital of Dalian University of Technology Dalian China Faculty of Information Technology University of Jyväskylä Jyväskylä Finland

Artificial intelligence and big data have great promise for promoting sleep medicine assistance. However, obstacles such as data privacy restrictions occur when enough data are collected from different data centers to develop classification models with better performance. Federated learning (FL), a collaborative machine learning approach, aims to address problems by acquiring models rather than directly accessing data from multiple clients. In this work, we utilize federated learning to increase model performance and preserve data privacy when deploying automatic sleep staging methods with insufficient data. And we construct a CNN-Transformer model and design an anti-class ratio (ACR) weighted loss function to complete the classification task with class imbalanced data. We evaluate our model and ACR using Sleep-EDF database and achieve better overall and N1 classification performance. To clearly show the efficiency of FL, we set three different data volume experiments with three independent heterogeneous public sleep databases. The experimental results are compared with two benchmark methods, traditional local learning and data-centralized learning. The results suggest that federated learning method outperforms the local learning approach in classification performance and achieves comparable results with data-centralized learning. As a result, we argue that federated learning is a promising alternative for multiple clients developing sleep staging algorithms especially with insufficient data because it adds benefits in terms of model performance and data privacy protection. © 2025 Elsevier Ltd

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Vesicle fusion and release in neurons under dynamic mechanical equilibrium

引用

iScience 2024年第5期27卷 109793页

作者： Liu, Wenhao Gao, Tianyu Li, Na Shao, Shuai Liu, Bo Cancer Hospital of Dalian University of Technology Shenyang 110042 China Faculty of Medicine Liaoning Key Lab of Integrated Circuit and Biomedical Electronic System Dalian University of Technology Dalian 116024 China

Vesicular fusion plays a pivotal role in cellular processes, involving stages like vesicle trafficking, fusion pore formation, content release, and membrane integration or separation. This dynamic process is regulated by a complex interplay of protein assemblies, osmotic forces, and membrane tension, which together maintain a mechanical equilibrium within the cell. Changes in cellular mechanics or external pressures prompt adjustments in this equilibrium, highlighting the system's adaptability. This review delves into the synergy between intracellular proteins, structural components, and external forces in facilitating vesicular fusion and release. It also explores how cells respond to mechanical stress, maintaining equilibrium and offering insights into vesicle fusion mechanisms and the development of neurological disorders. © 2024 The Authors

关键词： Biophysics Neuroscience

来源：评论

学校读者我要写书评

暂无评论

Analysis of a Low-Power Full Adder and Half Adder Using a New Adiabatic Logic 26

Analysis of a Low-Power Full Adder and Half Adder Using a Ne...

引用

26th International Conference on Computer and Information technology, ICCIT 2023

作者： Ganguly, Chamak Meem, Mazeda Zafor Kifayat Kabir, S.M. Biswas, Satyendra N. Primeasia University Dept. of EEE Dhaka Bangladesh Ahsanullah University of Science and Technology Dept. of EEE Dhaka Bangladesh ULKASEMI Pvt. Limited Circuit and System Design Dhaka Bangladesh Bloomsburg University Dept.of EEE AUST Dhaka Bangladesh and Dept. of BiCEP PA17815 United States

ISBN: (纸本)9798350359015

In today's fast-paced technological era, electronic devices must be both speedy and energy-efficient to keep up with rapid advancements. Developing energy-efficient digital devices for real-time applications is critical in VLSI research. The adiabatic process of logic circuits is used to develop an adder with high performance and ultra-low power dissipation. Techniques like multi-threshold devices (MTCMOS), variable multi-threshold devices (VTCMOS), and dynamic threshold devices (DTCMOS) have been explored with the scaling of devices. The proposed circuit contains a new type of adiabatic circuit that works excellently for low-power dissipation. A comparison is made with the conventional CMOS adders, such as half and full adders and newly developed adiabatic logic circuits. All the experimental simulations have been carried out using LTSpice software for 16nm and 22nm LP PTM models by changing frequencies from 0MHz to 25MHz, taking rise and fall time as 1n & 2n seconds respectively. We have also varied the load capacitance from 0f to 10f. The outcomes obtained from the experiment indicate the effectiveness and capability of the newly proposed circuit, which is 40% less power consumption and 22% less energy consumption for both 16nm and 22nm, respectively, using 0f as load capacitance. © 2023 IEEE.

关键词： Adiabatic CMOS Full adder Half adder Low power circuit power dissipation VLSI

来源：评论

学校读者我要写书评

暂无评论

Camera-based climbing analysis for a therapeutic training system

Camera-based climbing analysis for a therapeutic training sy...

引用

作者： Richter, Julia Beltrán, Raul Beltrán Heinkel, Ulrich Circuit and System Design Chemnitz University of Technology Chemnitz Germany

In view of therapeutic applications, climbing motion analysis has gained increased importance to avoid movements that are prone to cause injuries and to motivate the climber by means of gamification. To date, there remains need to investigate analysis methods for feedback generation that do not require body contact and that can be easily integrated in the climbing setup. Therefore, this study proposes an camera-based approach for contact-less motion analysis that localises the climber's centre of mass (COM) and derives relevant parameters, such as fluency, force and distance to the wall, from the temporal COM analysis. © 2020 Julia Richter et al., published by De Gruyter, Berlin/Boston 2020.

关键词： Computer vision

来源：评论

学校读者我要写书评

暂无评论

A novel semi-active wearable UHF RFID tag antenna for armpit temperature measurement 11

A novel semi-active wearable UHF RFID tag antenna for armpit...

引用

11th IEEE Asia-Pacific Conference on Antennas and Propagation, APCAP 2023 - Proceedings

作者： Liu, Bofu Zhou, Junyu Liu, Zixuan Yao, Minfeng Liu, Hui Guangdong Polytechnical Normal University School of Electronic and Information Guangzhou510665 China South China Normal University Rf Circuit & Microwave System Lab Centre for Optical and Electromagnetic Research Academy of Advanced Optoelectronics Guangzhou510006 China Guangdong University of Foreign Studies School of Information Science and Technology Guangzhou510006 China

ISBN: (纸本)9798350326277

In this paper, a semi-active wearable UHF RFID tag antenna with 'T' shape is proposed for measuring human armpit temperature. Both flexible polyimide and bio-silicone membrane material are used for tag substrate to obtain a good biocompatible between the tag and the human body. Simulated result of -10 dB impedance bandwidth of the tag is about 120 MHz, from 840 MHz to 960 MHz, which can cover the entire operating bandwidth of the global UHF RFID system. © 2023 IEEE.

关键词： temperature measurement tag UHF RFID antenna Wearable tag

来源：评论

学校读者我要写书评

暂无评论

An inexact alternating proximal gradient algorithm for nonnegative CP tensor decomposition

引用

Science China(Technological Sciences) 2021年第9期64卷 1893-1906页

作者： WANG DeQing CONG FengYu School of Biomedical Engineering Faculty of Electronic Information and Electrical EngineeringDalian University of TechnologyDalian 116024China Faculty of Information Technology University of JyväskyläJyväskylä40100Finland School of Artificial Intelligence Faculty of Electronic Information and Electrical EngineeringDalian University of TechnologyDalian 116024China Key Laboratory of Integrated Circuit and Biomedical Electronic System Liaoning ProvinceDalian University of TechnologyDalian 116024China

Nonnegative tensor decomposition has become increasingly important for multiway data analysis in recent years. The alternating proximal gradient(APG) is a popular optimization method for nonnegative tensor decomposition in the block coordinate descent framework. In this study, we propose an inexact version of the APG algorithm for nonnegative CANDECOMP/PARAFAC decomposition, wherein each factor matrix is updated by only finite inner iterations. We also propose a parameter warm-start method that can avoid the frequent parameter resetting of conventional APG methods and improve convergence *** experimental tests, we find that when the number of inner iterations is limited to around 10 to 20, the convergence speed is accelerated significantly without losing its low relative error. We evaluate our method on both synthetic and real-world *** results demonstrate that the proposed inexact APG algorithm exhibits outstanding performance on both convergence speed and computational precision compared with existing popular algorithms.

关键词： tensor decomposition nonnegative CANDECOMP/PARAFAC block coordinate descent alternating proximal gradient inexact scheme

来源：评论

学校读者我要写书评

暂无评论

Design and Implementation of Configurable UART with Self-Test Function

Design and Implementation of Configurable UART with Self-Tes...

引用

International Conference on Electronic Engineering and Informatics (EEI)

作者： Yuelun Zhang Zongguang Yu Xiaogang Yang China Electronics Technology Group Corporation No.58 Research Institute Wuxi China China Key System & Integrated Circuit Co. Ltd Wuxi China

Universal Asynchronous Receiver/Transmitter, commonly known as UART, is a configurable single-bit serial transmit and receive communication interface, which supports various baud rates, optional 5–8 bits of data length, parity check and optional 1–2 bits of stop bit. With the rapid development of MCU in the field of automotive electronics, the MCU that meets the requirements of vehicle regulations should have high reliability and efficient fault detection, and the application of the self-test function in UART can effectively improve the reliability and detection efficiency. The design is based on the self-test function and UART protocol design, using finite machine control timing, allowing the self-test controller to complete the self-test without manual configuration after receiving the self-test signal. The RTL of the UART and the self-test function are developed in Verilog, and the simulation verification is done in NCVerilog.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Novel 1T-APS Using 22 nm FD-SOI technology with Record High Tunable Sensitivity Range

IEEE Electron Device Letters

引用

IEEE Electron Device Letters 2025年

作者： Xie, Hui Wang, Zehua He, Rongshan Wang, Haihua Cao, Jingya Zhou, Peng Tang, Xiao-Lan Jiang, Yu-Long Xu, Yong Wan, Jing Fudan University College of Integrated Circuits and Micro-Nano Electronics Shanghai200433 China Nanjing University of Posts and Telecommunications College of Integrated Circuit Science and Engineering Nanjing210023 China Shaoxin Laboratory Shaoxing312000 China Gannan University of Science and Technology School of Information Engineering Ganzhou341000 China Guangdong Greater Bay Area Institute of Integrated Circuit and System Guangzhou510300 China

A novel one transistor active pixel sensor (1T-APS) with extraordinary tunable sensitivity range is experimentally demonstrated using 22 nm fully depleted silicon-on-insulator (FD-SOI) technology. The proposed device enables independent control of both full well capacity and sensitivity. Two electrodes are connected to the substrate, with one used to form the depletion region beneath the buried oxide (BOX) and the other to tune the movement of photoelectrons. The experimental results show a record high tunable sensitivity range of 340000%, varying from 4 mV/(μJ/cm2) to 13.6 V/(μJ/cm2). With tunable sensitivity, the dynamic range is expanded from 51.5 dB to 101.5 dB, making it promising for high-dynamic-range imaging applications. © 1980-2012 IEEE.

关键词： Pixels

来源：评论

学校读者我要写书评

暂无评论

A Wide-Band GaN Power Amplifier with Negative Feedback

A Wide-Band GaN Power Amplifier with Negative Feedback

引用

IEEE International Conference on Computer and Communication Engineering technology (CCET)

作者： Yimeng Fan YueWei Guo Lei Duan The 13th Research Institute China Electronics Technology Group Shijiazhuang China Circuit and System Department Bowei Integrated Circuits Company Ltd Shijiazhuang China

A two-stage common source wideband power amplifier with negative feedback network is presented in this paper. The design uses a negative feedback structure to improve the bandwidth and stability, and the matching network is also designed elaborately to meet the broadband requirements. The theoretical analysis of broadband design based on the parasitic effect of transistors is introduced and carried out in details in the paper. The bandwidth expanding technique with negative feedback structure are also analyzed. The dedicated theory analysis method is different from the impedance matrix theoretical wideband analysis in other designs. The proposed power amplifier designed based on a 0.25µm GaN process, has a saturated output power greater than 26dBm in the 8GHz-18GHz wide operating band, a maximum dynamic current of less than 149mA, and has a compact size of 1.7mm×1.1mm.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Single-Crystalline Linbo3 Integrated Onto Si-Based Substrates Via Surface Activation Bonding at Room Temperature

SSRN

引用

SSRN 2023年

作者： Huang, Rui Kan, Wanyu Tang, Mingzhi Xu, Hao Wu, Kai Li, Hui Wang, Qing Guo, Yecai Wang, Zhiyong Jiangsu Province Engineering Research Center of Integrated Circuit Reliability Technology and Testing System Wuxi university Wuxi214105 China Institute of Advanced Technology on Semiconductor Optics & Electronics Beijing University of Technology Beijing100124 China

LiNbO3 crystal is a multifunctional material with excellent performance and is widely used in integrated acousto-optic devices. Surface activated bonding enables the integration of two or more different materials at room temperature. In this paper, 4-inch LiNbO3/Si and LiNbO3/SiO2/Si bonded pairs were obtained at room temperature by optimizing Ar ion beam activation. After the pre-bonding was completed, a slicer was utilized to cut the LiNbO3/Si and LiNbO3/SiO2/Si pairs into 10 mm × 10 mm squares, respectively. The optimal annealing temperature is determined by multiple annealing experiments. Scanning acoustic microscope was used to confirm the high bonding rate of the two bonding pairs. Based on hydrophilic experiments and atomic force microscopy, the changes of hydrophilicity and roughness of LiNbO3, Si and SiO2 surfaces before and after activation can be compared. X-ray photoelectron spectroscopy was used to characterize chemical structure composition of samples. The dense interface without defects was observed by transmission electron microscopy. In addition, we explained the bonding mechanism between LiNbO3 and Si-based materials. The reasons for the different bonding strength of LiNbO3 with Si and SiO2 were also analyzed. Finally, the high bonding quality of LiNbO3 and Si-based materials can meet the stringent material requirements of Si-based LiNbO3 devices. © 2023, The Authors. All rights reserved.

关键词： Niobium compounds

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：