检索结果-内蒙古大学图书馆

11th IEEE International Conference on Power Electronics, Drives and Energy Systems, PEDES 2024

作者： Sawant, Rajendra R. Chaudhari, Bhalchandra N. Rao, Yerramreddy Srinivasa Sankhe, Darshana N. Padole, Anshal S. Sardar Patel Institute of Technology Dept. of Electronics & Telecomm. Mumbai400058 India D. J. Sanghvi College of Engineering Dept. of Electronics & Telecomm. Mumbai400056 India

The exponential growth in smartphone usage over GSM networks has significantly increased the energy demands of expanding telecom infrastructure. Concurrently, the adoption of green energy from renewable sources has become a key trend in the telecom industry. This paper presents the design and implementation of a hybrid PV-solar/Grid powered Telecom Power Plant (TPP) suitable for operation at modern roof-top mobile base stations. The conventional grid-powered telecom hardware is efficiently modified to operate on PV-solar power during daylight hours. The proposed TPP supports three rectifiers, each with a 2 kW capacity giving overall power of 6 kW, connected in parallel with hot-plugging features for redundancy, flexibility, and ease of maintenance. The power supply automatically switches between the PV source and the grid supply, based on the available voltage at the PV panels. The Power Factor Correction (PFC) stage in grid-mode is optimized for PV integration using Maximum Power Point Tracking (MPPT) control via a synchronous boost converter. The system employed by PFC stage, followed by an LLC resonant converter topology for a 400V to 54V DC-DC converter with synchronous rectifiers at the output. A multi-mode digital control system, implemented on a DSP controller, regulates the output voltage and current. This system utilizes pulse frequency modulation to meet the dynamic energy requirements of the telecom load over a wide current range, including boost-charging of the battery in Constant Current/Constant Voltage mode. A prototype 2 kW system was designed and implemented using Silicon Carbide power MOSFETs for most power switches. It was tested with a 3.3 kWp polycrystalline PV solar panel array and a 100Ah 48V Sealed Maintenance Free battery pack, supporting a 48V/60A 5G mobile-tower base-station. The proposed system operates at around 94% efficiency, utilizing an average 12-15 kWh of green energy per day. © 2024 IEEE.

关键词： telecomm.nication industry

来源：评论

学校读者我要写书评

暂无评论

Simultaneous Localization and Mapping (SLAM) in Swarm Robots for Map-Merging and Uniform Map Generation Using ROS 15

Simultaneous Localization and Mapping (SLAM) in Swarm Robots...

引用

15th International Conference on computer and Automation engineering, ICCAE 2023

作者： Salunke, Abhishek Patil, Chinmay Mude, Ram Joshi, Radhika D. College of Engineering Pune Electronics and Telecomm. Dept. Pune India

ISBN: (纸本)9798350396225

Swarm robotics is a sort of multi-robotics where many robots work together to coordinate in a distributed and decentralized way. It is based on the application of local rules and basic robots relative to the difficulty of the task, and it is motivated by social insects. A collection of basic robots with greater resilience and flexibility than a single robot may be able to do challenging tasks more effectively. In this paper, a multi-robot system is proposed- Swarm Robots for Simultaneous Localization And Mapping (SLAM). Individual robots in the swarm interact with each other and contribute to a single map as it localizes itself in the environment. The robots are equipped with RPLIDAR which is utilised for the hardware implementation of SLAM. The map constructed by each robot in the swarm is merged to make a uniform map. This map can be used by any other robot to effectively traverse an unknown environment and perform tasks that include search and rescue, cleanup of toxic spills, collective transportation, and unknown environment exploration like areas where it is dangerous for humans to work. © 2023 IEEE.

关键词： Multipurpose robots

来源：评论

学校读者我要写书评

暂无评论

An Efficient Video Summarization Using Sparse Auto-encoders and K-means Clustering on Video Similarity Graph 22

An Efficient Video Summarization Using Sparse Auto-encoders ...

引用

22nd OITS International Conference on Information Technology, OCIT 2024

作者： Kumari, T. Mita Dash, Dinesh Kumar Sahu, Abhimanyu Dept. of Electronics & Telecomm. Engg. Biju Patnaik University of Technology Odisha Rourkela India Dept. of Electronics & Telecomm. Engg. Parala Maharaja Engineering College Odisha Berhampur India Dept. of Computer Sc. & Engg. MNNIT Allahabad Prayagraj India

ISBN: (纸本)9798331510404

This paper presents a novel approach for video summarization integrating sparse auto-encoders and K-means clustering based on the video similarity graph (VSG). The suggested approach begins by separating each frame from the input video. Then, each frame is augmented by incorporating some minor changes. Sparse auto-encoders (SAE) are used to create a high-level feature map of every frame following augmentation. Afterwards, K-means clustering based on the video similarity graph of feature vectors is used to create the clusters so that each cluster has frames with similar features. Finally, the most informative video summary is created by extracting key-frames from each cluster. We tested the proposed method on a single standard benchmark SumMe dataset, and the findings undeniably show how much better the given approach is than several state-of-the-art techniques. © 2024 IEEE.

关键词： K-means clustering

来源：评论

学校读者我要写书评

暂无评论

Phydi: Initializing Parameterized Hypercomplex Neural Networks As Identity Functions 33

Phydi: Initializing Parameterized Hypercomplex Neural Networ...

引用

33rd IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2023

作者： Mancanelli, Matteo Grassucci, Eleonora Uncini, Aurelio Comminiello, Danilo Sapienza University of Rome Dept. of Information Engineering Electronics and Telecomm. Italy

ISBN: (纸本)9798350324112

Neural models based on hypercomplex algebra systems are growing and prolificating for a plethora of applications, ranging from computer vision to natural language processing. Hand in hand with their adoption, parameterized hypercomplex neural networks (PHNNs) are growing in size and no techniques have been adopted so far to control their convergence at a large scale. In this paper, we study PHNNs convergence and propose parameterized hypercomplex identity initialization (PHYDI), a method to improve their convergence at different scales, leading to more robust performance when the number of layers scales up, while also reaching the same performance with fewer iterations. We show the effectiveness of this approach in different benchmarks and with common PHNNs with ResNets- and Transformer-based architecture. The code is available at https://***/ispamm/PHYDI. © 2023 IEEE.

关键词： Parameterization

来源：评论

学校读者我要写书评

暂无评论

An Efficient Video Summarization Technique Using Texture Feature and Spectral Clustering 21

An Efficient Video Summarization Technique Using Texture Fea...

引用

21st OITS International Conference on Information Technology, OCIT 2023

作者： Mita Kumari, T. Kumar Dash, Dinesh Sahu, Abhimanyu Biju Patnaik University of Technology Dept. of Electronics & Telecomm. Engg. Odisha Rourkela India Parala Maharaja Engineering College Dept. of Electronics & Telecomm. Engg. Odisha Berhampur India MNNIT Dept. of Computer Sc. & Engg. Prayagraj Allahabad India

ISBN: (纸本)9798350358230

In this paper, a novel approach for video summarization combining spectral clustering and texture feature extraction is presented. Local Optimal Oriented Pattern (LOOP) textural feature extraction technique is explored for each frame of the video in our method for extracting features. Then, a clustering algorithm based on spectral clustering technique is applied to obtain an informative and an efficient video summary by extracting interesting key-frames. We evaluate our an efficient video summarization technique on one standard benchmark SumMe dataset and result clearly illustrates the preeminence of the suggested model in comparison to some of state-of-the-art methods.. © 2023 IEEE.

关键词： Clustering algorithms

来源：评论

学校读者我要写书评

暂无评论

Machine Monitoring for Industry using computer Vision

Machine Monitoring for Industry using Computer Vision

引用

2022 IEEE Conference on Interdisciplinary Approaches in Technology and Management for Social Innovation, IATMSI 2022

作者： Chiwande, Sujata S. Meshram, Piyush Charde, Abhishek Bhave, Shreya Nagdeote, Sushma Yeshwantrao Chavan College of Engineering Electronics & Telecomm. Dept. Nagpur India Fr. Conceicao Rodrigues College of Engineering Dept. of Computer Engineering Mumbai India

ISBN: (纸本)9781665477192

There are numerous approaches from which computer vision has been investigated. It moves beyond simply recording raw data to incorporate methods and concepts for computer graphics, pattern detection, digital image processing, and machine learning. This paper gives an outline of current technological advancements and theoretical ideas that describe how computer vision, mainly relates to image processing, and how it has evolved through time. It uses a technique for large-scale data analysis and a variety of application domains. The various research papers on computer vision and different techniques on object detection are reviewed in this paper. This paper gives the application of computer vision for factory and machine monitoring which will help to detect the object is moving or stable using YOLO algorithm. We also give a succinct summary of the most recent data regarding the effectiveness of the strategies. © 2022 IEEE.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

Adaptive Traffic Signal Timing: Leveraging YOLOv10 and computer Vision for Real-Time Optimization 8

Adaptive Traffic Signal Timing: Leveraging YOLOv10 and Compu...

引用

8th IEEE International Conference on Computational System and Information Technology for Sustainable Solutions, CSITSS 2024

作者： Papanashi, Suhas Chadaga, Manya Kshithi, R. Huddar, Santosh S. Sreelakshmi, K. Ramakanth Kumar, P. Rv College of Engineering Dept. of Electronics and Telecomm Bengaluru India Rv College of Engineering Dept. of Computer Science & Eng. Bengaluru India

ISBN: (纸本)9798331505462

Traffic congestion poses a significant challenge in modern urban environments, particularly in densely populated cities. Traditional traffic signals, which operate on fixed timings, often require manual adjustments to respond to varying traffic densities, leading to inefficient traffic flow. This issue arises because fixed intervals do not account for the dynamic nature of traffic patterns throughout the day. While developing intelligent systems that adjust signal durations dynamically offers a potential solution, such systems involve substantial financial investment and extended implementation timelines. This paper proposes a method to optimize green light durations by analyzing footage collected from a traffic junction in Bengaluru. This analysis can be performed using existing CCTV footage or manually gathered data. Even limited sample sizes can provide valuable insights for preliminary adjustments. The approach utilizes YOLOv10, a state-of-the-art computer vision algorithm, to extract relevant traffic data. The data is processed by an algorithm designed to reallocate green light durations, which are then compared with the original timings. An experimental setup involved recording one-minute videos over three days at the selected junction and applying the proposed algorithm to assess and refine signal durations. The results demonstrate the feasibility of using such algorithms to improve traffic signal management effectively. © 2024 IEEE.

关键词： Investments

来源：评论

学校读者我要写书评

暂无评论

GPU Reliability Assessment: Insights Across the Abstraction Layers

GPU Reliability Assessment: Insights Across the Abstraction ...

引用

2024 IEEE International Conference on Cluster Computing, CLUSTER 2024

作者： Yang, Lishan Papadimitriou, George Sartzetakis, Dimitris Jog, Adwait Smirni, Evgenia Gizopoulos, Dimitris George Mason University Department of Computer Science United States University of Athens Dept. of Informatics & Telecomm. Greece University of Virginia Department of Computer Science United States William & Mary Department of Computer Science United States

ISBN: (纸本)9798350358711

Graphics Processing Units (GPUs) are widely de-ployed and utilized across various computing domains including cloud and high-performance computing. Considering its extensive usage and increasing popularity, ensuring GPU reliability is cru-cial. Software-based reliability evaluation methodologies, though fast, often neglect the complex hardware details of modern GPU designs. This oversight could lead to misleading measurements and misguided decisions regarding protection strategies. This paper breaks new ground by conducting an in-dept. examination of well-established vulnerability assessment methods for modern GPU architectures, from the microarchitecture all the way to the software layers. It highlights divergences between popular software-based vulnerability evaluation methods and the ground truth cross-layer evaluation, which persist even under strong protections like triple modular redundancy. Accurate evaluation requires considering fault distribution from hardware to software. Our comprehensive measurements offer valuable insights into the accurate assessment of GPU reliability. © 2024 IEEE.

关键词： Program processors

来源：评论

学校读者我要写书评

暂无评论

Design of First Order Cascadable All Pass Filter Using FDCCII

Design of First Order Cascadable All Pass Filter Using FDCCI...

引用

2023 IEEE International Conference on Next Generation Electronics, NEleX 2023

作者： Kumar, Ashok Kushwaha, Ajay Kumar National Institute of Technology Dept. of Elec. and Comm. Engg Srinagar India College of Engineering Dept. of Elec. &Telecomm. Engg Pune India

ISBN: (纸本)9798350319088

A single fully differential second-generation current conveyor (FDCCII) and grounded passive components, as well as two new cascadable voltage mode (VM) first-order all-pass (AP) filters, are proposed in this study. From a cascading point of view for voltage-mode circuits, all of the presented designs have the characteristics of higher input and lower output impedances. To evaluate the effectiveness of the filters and confirm the proposed theory, simulation results using the SPICE tool are used. © 2023 IEEE.

关键词： All-pass filters

来源：评论

学校读者我要写书评

暂无评论

GRAMIAN MULTIMODAL REPRESENTATION LEARNING AND ALIGNMENT

arXiv

引用

arXiv 2024年

作者： Cicchetti, Giordano Grassucci, Eleonora Sigillo, Luigi Comminiello, Danilo Dept. of Information Engineering Electronics and Telecomm. Sapienza University of Rome Italy

Human perception integrates multiple modalities—such as vision, hearing, and language—into a unified understanding of the surrounding reality. While recent multimodal models have achieved significant progress by aligning pairs of modalities via contrastive learning, their solutions are unsuitable when scaling to multiple modalities. These models typically align each modality to a designated anchor without ensuring the alignment of all modalities with each other, leading to suboptimal performance in tasks requiring a joint understanding of multiple modalities. In this paper, we structurally rethink the pairwise conventional approach to multimodal learning and we present the novel Gramian Representation Alignment Measure (GRAM), which overcomes the above-mentioned limitations. GRAM learns and then aligns n modalities directly in the higher-dimensional space in which modality embeddings lie by minimizing the Gramian volume of the k-dimensional parallelotope spanned by the modality vectors, ensuring the geometric alignment of all modalities simultaneously. GRAM can replace cosine similarity in any downstream method, holding for 2 to n modalities and providing more meaningful alignment with respect to previous similarity measures. The novel GRAM-based contrastive loss function enhances the alignment of multimodal models in the higher-dimensional embedding space, leading to new state-of-the-art performance in downstream tasks such as video-audio-text retrieval and audio-video classification. The project page, the code and the pretrained models are available at https://***/GRAM/. © 2024, CC BY.

关键词： Contrastive Learning

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：