检索结果-内蒙古大学图书馆

A novel human action recognition using Grad-CAM visualization with gated recurrent units

Neural Computing and Applications 2025年 1-16页

作者： Jayamohan, M. Yuvaraj, S. Department of Electronics and Communication Engineering College of Engineering and Technology SRM Institute of Science and Technology Tamilnadu Kattankulathur603203 India

Human action recognition is a vital aspect of computer vision, with applications ranging from security systems to interactive technology. Our study presents a comprehensive methodology that employs multiple feature extraction and optimization techniques to enhance the accuracy and efficiency of human action identification. The video input was divided into four distinct elements: RGB images, optical flow information, spatial saliency maps, and temporal saliency maps. Each component was analyzed independently using advanced computer vision algorithms. The process involves utilizing various algorithms and techniques to extract meaningful information from the visual data. The Farneback algorithm was employed to examine the optical flow, whereas Canny edge detection was used to assess spatial prominence. Additionally, frame comparison helps to identify motion-based prominence. These processed elements provide a comprehensive representation of both spatial and temporal information. The extracted data were then input into distinct pretrained deep learning models. Specifically, Inception V3 was used for RGB frames and optical flow analysis, ResNetV2 processed spatial saliency maps, and DenseNet-121 handled motion saliency maps. The input data are processed separately by these networks, each of which extracts specific features that are suited to their respective modalities. This feature extraction process ensures the comprehensive capture of both static and dynamic elements in video data. Subsequently, sequence modeling and classification were performed using a gated recurrent unit (GRU) that incorporated an attention mechanism. This mechanism dynamically highlights the most significant temporal segments, improving the capacity of the model to comprehend intricate human actions within video sequences. To enhance the efficiency of the model, we implemented the Grasshopper optimization algorithm to optimize the feature selection and classification stages, thus maximizing the u

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

UAV-Assisted communications With RF Energy Harvesting: A Comprehensive Survey

引用

IEEE communications Surveys and Tutorials 2025年第2期27卷 782-838页

作者： Pandey, Gaurav Kumar Gurjar, Devendra Singh Yadav, Suneel Jiang, Yuming Yuen, Chau National Institute of Technology Silchar Department of Electronics and Communication Engineering Silchar788010 India Indian Institute of Information Technology Allahabad Department of Electronics and Communication Engineering Prayagraj211015 India Department of Information Security and Communication Technology Trondheim7491 Norway Nanyang Technological University School of Electrical and Electronic Engineering Nanyang639798 Singapore

Unmanned aerial vehicles (UAVs) have recently achieved sky-rocketed prominence in assisting various wireless communication scenarios due to their implicit characteristics like line-of-sight connectivity, three-dimensional mobility, on-demand deployment, and payload abilities. For realizing UAV-assisted communications in practical scenarios, one of the key challenges is the onboard power limitation of UAVs, which can affect their flight duration, maneuverability, and network lifetime. Exploiting energy harvesting (EH) techniques with UAV networks can improve operational time by addressing the UAV's power constraints. In addition, UAVs can facilitate seamless connectivity for energy-constrained low-power devices in remote locations, disaster-struck areas, emergencies, and areas where human intervention is challenging and recharging or replacing batteries is problematic. In such scenarios, a low-altitude UAV, which flies flexibly, can serve as a radio-frequency (RF) energy transmitter to charge the nearby low-power devices efficiently and help enable their communications. Focusing on these aspects, this paper comprehensively surveys the current state-of-the-art EH techniques that can be utilized with UAV communications in various practical scenarios. Then, we highlight crucial challenges in integrating UAVs with prevailing wireless infrastructure and discuss different RF-EH receiver architectures that can be viable solutions for power-constrained UAV networks. Further, we show the channel modeling of UAV-assisted communications with EH and provide critical insights into the system performance. Besides, we offer an exhaustive review of recent works on RF-EH in UAV networks with next-generation paradigms and different problem-solving techniques/optimization methods. Finally, we suggest forthcoming trends and highlight open issues and future research directions. © 1998-2012 IEEE.

关键词： Signal to noise ratio

来源：评论

学校读者我要写书评

暂无评论

KHOWmark: optimized dual image watermarking and its two-level authentication for smart healthcare system in schur domain

引用

Multimedia Tools and Applications 2025年 1-28页

作者： Tiwari, Anurag Awasthi, Divyanshu Srivastava, Vinay Kumar Electronics and Communication Engineering Department Motilal Nehru National Institute of Technology Uttar Pradesh Prayagraj211004 India School of Electrical and Electronics Engineering VIT BHOPAL University Sehore466114 India Department of Electronics and Communication Engineering GLA University Mathura MATHURA India

A wide solution to address the growing diversity of specialized security threats in healthcare information management, watermarking has been proposed in response to contemporary worries about multimedia security and the necessity of efficiently storing, accessing, and transmitting enormous volumes of medical data. Dual watermark-based methods can be utilized simultaneously for various image watermarking applications, however the real-time implementation of a single watermark-based method has a number of drawbacks. In this work, Lifting wavelet transform (LWT)-Discrete cosine transform-Schur decomposition (SD) based dual image watermarking scheme is proposed for medical applications in which zigzag scanning based selective embedding is done to provide balance between imperceptibility and robustness. In the proposed scheme, two watermark images: (i) patient ‘Aadhaar’ card and (ii) patient ‘fingerprint’ are taken to embed into the cover ultrasound image. In the presented work, embedding gain factor is optimized using the Krill herd optimization in order to balance imperceptibility and robustness. In this work, Advanced encryption standard authentication and features from accelerated segment test feature-based authentication is done to verify the watermarked image. The proposed method offers more than 10% improvement in performance parameters in most of the cases. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2025.

关键词： Authentication

来源：评论

学校读者我要写书评

暂无评论

Enhanced non-enzymatic multicomponent detection via one-step hydrothermal synthesis of widely dispersed Zn-SnO2 nanoparticles on nitrogen-doped reduced graphene oxide

引用

Materials Research Innovations 2025年第2期29卷 99-107页

作者： Baraneedharan, P. J Sephra, Percy Department Electronics and Communication Engineering Saveetha Engineering College Chennai India Department of Electronics and Communication Engineering B.S. Abdur Rahman Crescent Institute of Science and Technology Chennai India

Herein, we report a new type of non-enzymatic nanostructured biosensor designed for the concurrent determination of dopamine, cholesterol, and glucose in human blood samples. The prepared biosensor using low-temperature one-pot hydrothermal processing is a Zn-doped SnO2-dispersed nitrogen-doped reduced graphene oxide (N-rGO)/Zn-SnO2 nanocomposite. This modified electrode allows direct electron transfer and thus offers electrochemical techniques to detect target molecules in analytes without the use of enzymes. The biosensor displayed good voltammetric responses within the human reference range: 20–70 nM for dopamine, 1–8 mM for cholesterol, and 1–10 mM for glucose. It also showed excellent reusability and can be restored by washing with ethanol. Due to the use of Zn-doped SnO2 nanoparticles for enhanced catalytic activity and N-rGO for high conductivity and active sites, sensitive and selective detection can be achieved. This hybrid composite demonstrates excellent catalytic activity, repeatability, stability, and clinical comparability for real-time measurement of dopamine, cholesterol, and glucose simultaneously. © 2024 Informa UK Limited, trading as Taylor & Francis Group.

关键词： Cholesterol

来源：评论

学校读者我要写书评

暂无评论

Investigation of BTBT-Triggered Split-Gate Junctionless Nanosheet FETs for Analog and RF Applications

引用

Transactions on Electrical and Electronic Materials 2025年 1-13页

作者： Kumar, Subhash Kumar, Arun Srinivas, P.S.T.N. Department of Electronics and Communication Engineering National Institute of Technology Silchar India School of Electronics Engineering Vellore Institute of Technology Vellore India

This work investigates the performance of the Split-Gate Junctionless Nanosheet Field-Effect Transistor (SG-JL-NSFET), focusing on its suitability for analog and RF applications. The SG-JL-NSFET features an n-type channel region with uniform doping concentration and have two distinct gate regions. One of the gates towards the source end is the control gate, whereas the other gate is the floating gate, which has different work functions separated by a thin spacer region. This split-gate architecture triggers the band-to-band-tunneling (BTBT) action in the channel region, thereby improving the drive current and offering an improved on-off current ratio (I_ON/I_OFF) of 1.74 × 10¹⁵. Considering the SG-JL-NSFETs for radio frequency circuit applications, a thorough analysis of analog characteristics such as transconductance, gate capacitances, and cut-off frequency has also been undertaken in this work. We have also investigated the RF and linearity parameters of SG-JL-NSFETs, such as transconductance generation factor (TGF), transconductance frequency product (TFP), gain bandwidth product (GBP), transit time, VIP2, VIP3, IIP3, IMD3 points. © The Korean Institute of Electrical and Electronic Material Engineers 2025.

关键词： Analog BTBT Junctionless NSFETs Radio frequency Transconductance

来源：评论

学校读者我要写书评

暂无评论

CUSE: Copyright protection of UNESCO heritage images for safe cities environment

引用

Multimedia Tools and Applications 2025年 1-34页

作者： Awasthi, Divyanshu Khare, Priyank Srivastava, Vinay Kumar Department of Electronics and Communication Engineering GLA University Mathura Mathura India Department of Electronics and Communication Engineering Indian Institute of Information Technology Ranchi Jharkhand Ranchi India Department of Electronics and Communication Engineering Motilal Nehru National Institute of Technology Allahabad Uttar Pradesh Prayagraj India

Images, habits, rituals, beliefs, cultures, information, arts, crafts, music, and artefacts specific to a place are all included in heritage. The cultural, natural and mixed heritage of India is very diverse, complicated, and complex. The United Nations Educational, Scientific and Cultural Organization (UNESCO) designates various sites of countries as the world heritage sites. The proposed watermarking technique aims to provide the copyright protection to the UNESCO world heritage sites image of India (UWHSII). This proposed method uses the novel and efficient combination of redundant discrete wavelet transform (RDWT) and multi-resolution singular value decomposition (MSVD). In order to balance the robustness and imperceptibility of any watermarking scheme, the scaling factor is essential. Aquila optimization (AO), and fuzzy inference system (FIS) is used to get the optimum scaling factor. The comparative analysis of AO and FIS is also carried out. Internet of cultural things (IoCT)-based real-time authentication is carried out by using the geographical location of sites and secured hash algorithm. To evaluate the performance of the proposed scheme Average peak-signal-to-noise ratio (APSNR), Structural similarity index measure (SSIM), Degree of balance (BD), Universal image quality index (UIQI) and Normalized correlation coefficient (NCC) is used. The average percentage improvement in robustness is 8.96%, and imperceptibility is 54.51%. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2025.

关键词： Fuzzy inference

来源：评论

学校读者我要写书评

暂无评论

PriChain: Efficient Privacy-Preserving Fine-Grained Redactable Blockchains in Decentralized Settings

引用

Chinese Journal of electronics 2025年第1期34卷 82-97页

作者： Hongchen Guo Weilin Gan Mingyang Zhao Chuan Zhang Tong Wu Liehuang Zhu Jingfeng Xue School of Computer Science and Technology Beijing Institute of Technology School of Cyberspace Science and Technology Beijing Institute of Technology School of Computer and Communication Engineering University of Science and Technology Beijing

Recently, redactable blockchain has been proposed and leveraged in a wide range of real systems for its unique properties of decentralization, traceability, and transparency while ensuring controllable on-chain data redaction. However, the development of redactable blockchain is now obstructed by three limitations, which are data privacy breaches, high communication overhead, and low searching efficiency, respectively. In this paper, we propose PriChain, the first efficient privacy-preserving fine-grained redactable blockchain in decentralized settings. PriChain provides data owners with rights to control who can read and redact on-chain data while maintaining downward compatibility, ensuring the one who can redact will be able to read. Specifically, inspired by the concept of multi-authority attribute-based encryption, we utilize the isomorphism of the access control tree, realizing fine-grained redaction mechanism, downward compatibility, and collusion resistance. With the newly designed structure, PriChain can realize O（n） communication and storage overhead compared to prior O（n2） schemes. Furthermore, we integrate multiple access trees into a tree-based dictionary, optimizing searching efficiency. Theoretical analysis proves that PriChain is secure against the chosen-plaintext attack and has competitive complexity. The experimental evaluations show that PriChain realizes 10× efficiency improvement of searching and 100× lower communication and storage overhead on average compared with existing schemes.

关键词： Resistance Access control Data privacy Dictionaries Blockchains Encryption Complexity theory Usability Protection

来源：评论

学校读者我要写书评

暂无评论

Capacitive Sensing in High-speed Molecular communication System: A Non-invasive Interface for the Internet of Bio-Nano Things

引用

IEEE Internet of Things Journal 2025年第11期12卷 16711-16719页

作者： Huang, Xuewei Liu, Jiarui Lin, Lin Wen, Miaowen Gan, Wancheng Huang, Yu Guangzhou University Research Center of Intelligent Communication Engineering School of Electronics and Communication Engineering Guangzhou510006 China Tongji University College of Electronics and Information Engineering Shanghai201804 China School of Electronic and Information Engineering in South China University of Technology Guangdong Provincial Key Laboratory of Short-Range Wireless Detection and Communication Guangzhou510641 China

Molecular communication (MC) is a prominent technique within the Internet of Bio-nano Things (IoBNT), which aims to interconnect artificial and biological devices for pioneering healthcare applications. Despite its rapid advancements in theoretical study, the experimental development of MC is still in the infancy stage, and most existing testbeds require intruding hardware into the channel during the signal detection process, potentially harming the organism. Against this background, the non-invasive interface across the internal and external information exchange of organisms is of great necessity. Specifically, this work adopts a capacitive sensing mechanism to recover information from the sensed chemical signals. Our testbed demonstrates a notably higher transmission rate than previously reported non-invasive MC approaches, underscoring its potential to advance IoBNT research. © 2025 IEEE.

关键词： Testbeds

来源：评论

学校读者我要写书评

暂无评论

A Novel Underlay Metal Strip Loaded Doping-Less Heterojunction (GaSb/Si) TFET Biosensor for Autoimmune Disease Detection

引用

IEEE Transactions on Nanotechnology 2025年 24卷 239-248页

作者： Verma, Madhulika Agrawal, Sachin National Institute of Technology Electronics and Communication Engineering Department Delhi110036 India

In human being autoimmune diseases are caused by the immune system's attack on body tissues. Therefore, advanced diagnostic tools for their early and accurate detection is highly needed. This study introduces a new underlay metal strip loaded doping-less heterojunction (GaSb/Si) TFET biosensor (UMS-DL-HJ-TFETB) device with exceptional sensitivity and performance. Key design features include an underlay metal strip for improved tunnelling and the cavities are on the source region to achieve a peak drain current sensitivity of 6.7 × 1010 at k = 12 and Vgs = 0.45 V. With a cut-off frequency of 3.27 × 108 Hz and a response time of 496 ps, the proposed biosensor exhibits excellent RF performance. The device performance in detecting DNA charge densities ranging from ± 1 × 1011 cm-2 to ± 1 × 1012 cm-2 has also been studied. In addition, five non-uniform distributions which is caused by the steric hindrance effect have been optimized. A comparative analysis is also done for fair evaluation. The simulation results show that the proposed biosensor addresses the limitations of conventional methods, providing high sensitivity, rapid detection and reliable diagnostic accuracy for autoimmune diseases. © 2002-2012 IEEE.

关键词： Strip metal

来源：评论

学校读者我要写书评

暂无评论

Deep learning empowered channel estimation in massive MIMO: unveiling the efficiency of hybrid deep learning architecture

引用

Journal of Ambient Intelligence and Humanized Computing 2025年第2期16卷 375-390页

作者： Ranjan, Amish Sahana, Bikash Chandra Department of Electronics and Communication Engineering National Institute of Technology Bihar Patna India

Massive Multiple-Input Multiple-Output (MIMO) technology has changed the way wireless connectivity works and promises to make spectral efficiency better than ever before. Traditional methods, like Maximum Ratio Transmission (MRT) beamforming, have problems with big antenna arrays, channel estimation errors, and wireless channel variability. To get rid of these problems, the proposed model presents a novel deep learning method that combines Gated Recurrent Units (GRU) and Long Short-Term Memory (LSTM) networks. It uses both temporal and spatial relationships in channel data to make channel estimation and beamforming better in massive MIMO systems. The comparison results show the efficiency of the proposed method with respect to state-of-the art methods for channel estimation in massive MIMO. At the beginning of the study, massive MIMO technology is thoroughly evaluated, with a focus on both its benefits and drawbacks. We discuss the theoretical foundations of MRT beamforming and its limitations when dealing with large antenna arrays. To tackle these challenges, we describe a novel deep learning architecture that leverages the temporal and spatial relationships seen in the channel data through the use of GRU and LSTM layers. A comprehensive method including model designs, training schedules, metrics for performance assessment, and data generation is explained. We perform comprehensive controlled simulations that allow us to compare the GRU + LSTM approach and MRT’s spectrum efficiency. The results offer exciting new insights. This research not only shows improved spectral efficiency and robustness to channel variations, but it also elucidates the trade-offs between deep learning and conventional methods in wireless communication systems, indicating that deep learning could be essential to achieving the full benefits of Massive MIMO. © The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2025.

关键词： Beamforming

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：