检索结果-内蒙古大学图书馆

UPB Scientific Bulletin, Series C: Electrical engineering and Computer Science 2021年第3期83卷 143-154页

作者： Diaconescu, Paul Neagoe, Victor-Emil Dept. of Applied Electronics and Information Engineering University POLITEHNICA of Bucharest Romania

This paper is dedicated to road object detection in urban environment using Deep Learning Neural Networks (DLNN). We have chosen the advanced architecture YOLOv5 trained on COCO 2017 dataset (containing 80 classes of objects in 1.5 million images) and we have specialized YOLOv5 to recognize objects from BDD100K dataset (Berkeley Diverse Driving Video Database, including 100,000 images). We have kept eight classes from BDD100K: car, truck, person, traffic sign, traffic light, bus, rider and bike. The object detection performance (average precision) obtained in this paper for BDD100K dataset is the highest, compared with previous published approaches for the same dataset, and the highest for the mapping of YOLOv5 and BDD100K dataset. The research results prove that one can apply state of the art cutting edge deep learning technologies to open new perspectives for automotive industry. © 2021, Politechnica University of Bucharest. All rights reserved.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Performance Analysis and Implementation of LSTM and GRU based on Synthetic Data 2

Performance Analysis and Implementation of LSTM and GRU base...

引用

2nd IEEE International Conference on Interdisciplinary Approaches in Technology and Management for Social Innovation, IATMSI 2024

作者： Rahul Annamalai, K. Jha, Arya Deepa, K. Tresa Sangeetha, S.V. Chaitanya, L. Channegowda, Janamejaya Amrita School of Engineering Amrita Vishwa Vidyapeetham Dept. of Electrical and Electronics Engineering Bengaluru India University of Technology and Applied Sciences Department of Engineering AlMusanna Oman Bms College of Engineering Department of Electrical and Electronics Engineering Bengaluru India L & T Technology Services Bengaluru India

ISBN: (纸本)9798350360523

The goal of this work is to develop artificial dataset through the use of autoencoders, a solution to the problem of expensive and scarce real-world data for predictive model training in the field of battery health monitoring. An autoencoder neural network is used to generate synthetic datasets that capture the complex patterns and features of battery behavior by utilizing a large dataset with 81,000 records. The scarcity of battery measurement data poses a challenge for battery researchers, as it can hinder the development of reliable estimation models. This paper addresses this issue by proposing machine learning-based techniques for the synthesis of high-fidelity battery datasets. The generated synthetic datasets can be employed to augment the existing datasets, thereby enhancing the effectiveness of the research endeavors. The presented work demonstrates that synthetic dataset can be used as the replica or as a substitute of the work done with larger dataset. The strategies presented in this paper offer a practical solution to overcome the limitations posed by small battery datasets. Moreover, a comparison has been provided between the ML method proving that LSTM emerges to be a better method based on the performance metrics with one drawback of having longer training time than GRU. © 2024 IEEE.

关键词： Large datasets

来源：评论

学校读者我要写书评

暂无评论

Effect of Gain Flattening Filter Placement for Nonlinear-Impairment Mitigation in Multiband Optical Transport Network 17

Effect of Gain Flattening Filter Placement for Nonlinear-Imp...

引用

17th IEEE International Conference on Advanced Networks and Telecommunications Systems, ANTS 2023

作者： Jana, Rana Kumar Srivastava, Anand Lord, Andrew Mitra, Abhijit Indraprastha Institute of Information Technology Dept. of Electronics and Communications Engineering New Delhi India British Telecom Applied Research United Kingdom Indraprastha Institute of Information Technology BITS-ON Group Dept. of Electronics and Communications Engineering New Delhi India

ISBN: (纸本)9798350307672

The multiband elastic optical technology appears as a near-term solution for solving the optical fiber capacity crunch problem in core optical networks. However, the significant challenges come from several non-linear physical-layer impairments (PLIs), such as inter-channel stimulated Raman scattering (ISRS), and band-dependent transmission impairments, such as attenuation, dispersion, etc. In order to maintain the quality of transmission (QoT) throughout all the bands, the gain flattening filters (GFFs) need to be placed in the network, which can be helpful to enhance the overall capacity of the network while mitigating the effect of PLIs. This paper investigates the effect of GFF placement at different parts of geographically diverse networks in terms of traffic admissibility, QoT, and return on investment (ROI). Reported results show that the placement of GFF at every inline amplifier module instead of its placement at the reconfigurable optical add-drop multiplexer (ROADM) module can provide a maximum gain of nearly 37.7% and 118.7% in traffic admissibility for smaller (such as BT-UK network) and larger networks (such as Indian RailTel networks), respectively. In addition, the effect of launch power and symbol rate variation is also studied in the paper. Numerical results show that the placement of GFF at the ROADM modules outperforms in terms of ROI, whereas the placement of GFF at the EDFA module outperforms in terms of QoT. © 2023 IEEE.

关键词： Erbium doped fiber amplifiers

来源：评论

学校读者我要写书评

暂无评论

A Shared Ground Economical Dielectric-Free Dual Polarized Antenna System for 5G Mobile Devices

A Shared Ground Economical Dielectric-Free Dual Polarized An...

引用

2023 IEEE International Symposium on Antennas and Propagation and USNC-URSI Radio Science Meeting, AP-S/URSI 2023

作者： Karthikeya, G.S. Koul, Shiban K. Poddar, A.K. Rohde, Ulrich Bms College of Engineering Dept. of Electronics and Communication India Center for Applied Research in Electronics Iit Delhi India Synergy Microwave Corporation NJ United States

ISBN: (纸本)9781665442282

A probe-fed dielectric-free antenna with wide impedance bandwidth is proposed in this paper. The conventional SMA (SubMiniature version A) connector could be used to realize this antenna. Here, the top plane is the radiating element with a modified cuboid. The ground plane is Copper with dimensions of 8mm x 8 mm. This dielectric-free antenna offers an impedance bandwidth of 12.2%. As the form factor is minimal, the antenna could be panel mounted with modern 5G mobile devices. The same radiator is extended for dual-polarized system with shared ground. © 2023 IEEE.

关键词： 5G mobile communication systems

来源：评论

学校读者我要写书评

暂无评论

An Ultra Low Cost Orthogonal Pattern Diversity Antenna Module for Compact mmWave 5G Mobile Devices

An Ultra Low Cost Orthogonal Pattern Diversity Antenna Modul...

引用

2023 IEEE International Symposium on Antennas and Propagation and USNC-URSI Radio Science Meeting, AP-S/URSI 2023

作者： Karthikeya, S.G. Koul, Shiban K Poddar, A.K. Rohde, Ulrich BMS College of Engineering Dept. of Electronics and Communication India IIT Delhi Center for Applied Research in Electronics India Synergy Microwave Corporation NJ United States

ISBN: (纸本)9781665442282

A wideband microstrip-fed end-fire antenna is proposed in this paper. The antenna is designed on the low-cost commercially viable FR4 substrate. The proposed antenna is electrically compact with dimensions, 2.1λ × 0.466 λ × 0.14 λ, the free space wavelength is calculated at 28GHz. The input impedance bandwidth is 46% with reasonably high pattern integrity across the operational bandwidth. A panel mountable two port pattern diversity system with low mutual coupling is also presented. © 2023 IEEE.

关键词： Costs

来源：评论

学校读者我要写书评

暂无评论

Speech based Emotion Recognition and Gender Identification using FNN and CNN Models 3

Speech based Emotion Recognition and Gender Identification u...

引用

3rd International Conference for Emerging Technology, INCET 2022

作者： Susithra, N. Rajalakshmi, K. Ashwath, P. Ajay, B. Rohit, D. Stewaugh, S. Psg Dept. of Electronics and Communication Engineering Institute of Technology and Applied Research Coimbatore India Psg Dept. of Electronics and Communication Engineering College of Technology Coimbatore India

ISBN: (纸本)9781665494991

Detection of human emotions by analyzing the speech patterns in the audio signals has various applications including effective human-computer interaction, evaluation or automation of the customer care call quality assurance process, and medical applications such as assisting a doctor in understanding the emotional state of a patient. For instance, speech produced in a state of anger, fear, or happiness has higher and wider pitch and it is louder and faster whereas slow and low-pitched speech is produced when a person is sad. Recognition of emotion, gender and age from speech signals is one of the active research areas in speech processing where there is need for improvement in accuracy. The work proposed in this paper makes use of machine learning neural networks to recognize the gender and emotion of a speaker. The model is designed to classify four different emotions (neutral, happy, sad, angry). The work involves the integration of two modules namely, Gender identification block and Emotion detection block. A simple Feed forward neural network, tested and trained by a large dataset forms the gender identification block, while Convolutional Neural Network (CNN) with adequate datasets were used to train and test the emotion detection block. The proposed model results in detection of gender with 91.46 percent accuracy while the emotion detection module provides 86 percent accuracy. © 2022 IEEE.

关键词： Large dataset

来源：评论

学校读者我要写书评

暂无评论

Routing in WSN Based on Improving LEACH Protocol

Routing in WSN Based on Improving LEACH Protocol

引用

2024 Multimedia University engineering Conference, MECON 2024

作者： Saoud, Bilal Shayea, Ibraheem Roslee, Mardeni University of Bouira Faculty of Sciences and Applied Sciences LISEA Laboratory Dept. Electrical Engineering Bouira10000 Algeria Electronics & Communications Engineering Faculty of Electrical and Electronics Engineering Istanbul34469 Turkey Centre for Wireless Technology Faculty of Engineering Multimedia University Cyberjaya63100 Malaysia

ISBN: (纸本)9798331530747

Wireless sensor networks (WSNs) are now essential for a range of applications, such as environmental monitoring, surveillance and industrial automation. An important obstacle in WSNs is the constrained energy resources of each sensor node, which greatly affects the overall lifespan of the network. It is essential to prolong the lifespan of the network in order to provide continuous and uninterrupted data collecting. This work introduces a novel routing protocol that builds upon the LEACH protocol. The implementation has been executed and verified by testing. The simulation findings demonstrate the efficacy of our novel proposition in relation to the LEACH protocol, specifically in terms of the longevity of the network and energy consumption. © 2024 IEEE.

关键词： Environmental monitoring

来源：评论

学校读者我要写书评

暂无评论

Design and Performance Study of LCC-LCC and LCC-S Compensation Network for Wireless Charging of EV Battery 5

Design and Performance Study of LCC-LCC and LCC-S Compensati...

引用

5th IEEE International Transportation Electrification Conference, ITEC-India 2023

作者： Balaji, C. Venugopal, R. Selvaraj, Jayachitra Balasundaram, G. Savio, A. Dominic Lim, Wei Hong Srm Institute of Science and Technology Dept. of Electrical and Electronics Engineering Chennai India University of Technology and Applied Science Eee Section Dept. of Electrical Power Division Muscat Oman Dept. of Electrical and Electronics Engineering Chittor RVS nagar Andra pradesh India Ucsi University Faculty of Engineering Technology and Built Environment Malaysia

ISBN: (纸本)9798350337808

This work aims to study the compensation techniques used in wireless power transfer for Electric Vehicle (EV) charging application. Though there are several topologies for wireless power conversion Dual Active Bridge (DAB) configuration, better suits with wireless power transfer for its bidirectionality, high power density, ease of implementation and so on. Compensation network enhances the performance of the DAB in providing the constant current (CC) or constant voltage (CV) according to the battery charging status. This paper details the study on the performance of LCC-LCC topology and LCC- S topologies. The design values of magnetic elements of the compensation network are calculated analytically. The LCC-LCC and LCC- S compensations implementation in DAB converter is simulated in MATLAB/Simulink. The results were obtained and compared with the symmetrical LCC and LCC- S compensation on the factors of efficiency, output power, transmitter current and receiver current. © 2023 IEEE.

关键词： Topology

来源：评论

学校读者我要写书评

暂无评论

BINARY PROGRAMMING MODELS FOR ENERGYEFFICIENT VIRTUAL MACHINES PLACEMENT IN DATA CENTERS

UPB Scientific Bulletin, Series C: Electrical Engineering an...

引用

UPB Scientific Bulletin, Series C: Electrical engineering and Computer Science 2024年第3期86卷 335-346页

作者： Radulescu, Delia Mihaela Radulescu, Delia Mihaela Radulescu, Marius Radulescu, Constanţa Zoie Lazaroiu, Gheorghe National Institute for Research and Development in Informatics Bucharest Romania Faculty of Power Engineering National University of Science and Technology POLITEHNICA Bucharest Romania Dept. Statistical Inference Statistical Control and Optimization Gheorghe Mihoc-Caius Iacob Institute of Mathematical Statistics and Applied Mathematics Romanian Academy Bucharest Romania Senior research worker National Institute for Research and Development in Informatics Bucharest Romania Prof. Dept. of Energy Production and Use National University of Science and Technology POLITEHNICA Bucharest Romania

The aim of the paper is to present two virtual machines (VMs) placement models that minimize energy consumption in a data center. Both placement models use binary programming. The first model aims to find an allocation of VMs to a set of servers such that the energy consumption is minimized. The second model suppose that an allocation of VMs to a set of servers is given. Its aim is to find another allocation of VMs to servers that minimize the sum of two energies. The first energy is the energy consumption of VMs in the new allocation and second energy is the energy needed for migration between the two allocations. © 2024, Politechnica University of Bucharest. All rights reserved.

关键词： allocation binary programming model data center energy consumption minimization servers virtual machines

来源：评论

学校读者我要写书评

暂无评论

TETI Project - A Multiparameter Modular Buoy for a Comprehensive and Cost Effective Sea Water Monitoring

TETI Project - A Multiparameter Modular Buoy for a Comprehen...

引用

2024 IEEE International Workshop on Metrology for Living Environment, MetroLivEnv 2024

作者： Panfiglio, Simone Trovato, Valentina Insolvibile, Gianluca Cannuli, Antonio Plutino, Maria Rosaria Iannucci, Leonardo Montanini, Roberto Sfameni, Silvia Lombardo, Luca Quattrocchi, Antonino De Santis, Luca Grassini, Sabrina University of Messina NAVTEC Messina Italy University of Bergamo Dept. of Engineering and Applied Sciences BG Dalmine Italy Nextworks Srl Pisa Italy University of Messina Dept. of Engineering Messina Italy ISMN CNR ME Messina Italy Politecnico di Torino Dept. Of Applied Science and Technology Turin Italy Politecnico di Torino Dept. of Electronics and Telecommunications Turin Italy Net7 Srl Pisa Italy

ISBN: (纸本)9798350385014

TETI is a multidisciplinary project aimed at developing innovative sensors to monitor different kind of seawater pollution, exploiting the possibility to integrate them completely or in part into a multiservice buoy, to be used in marinas and small ports to provide energy supply to small electrical boats giving at the same time useful information about the health status of the waters. The system is designed to send the desired set of information on shore using an integrated wireless system, interfaced with an intuitive user interface. The article aims to describe the overall modular architecture of the system, a part of the sensors developed for on- board integration, the power supply system and data transmission system. © 2024 IEEE.

关键词： Information use

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：