检索结果-内蒙古大学图书馆

IEEE Transactions on Artificial Intelligence 2024年第7期5卷 3579-3594页

作者： Chakraborty, Prabuddha Bhunia, Swarup University of Maine Department of Electrical & Computer Engineering OronoME04469 United States University of Florida Department of Electrical & Computer Engineering GainesvilleFL32603 United States

Edge computing devices in Internet-of-Things (IoT) systems are being widely used in diverse application domains including industrial automation, surveillance, and smart housing. These applications typically employ a large array of sensors, store a high volume of data, and search within the stored data for specific patterns using machine intelligence. Due to this heavy reliance on data in these applications, optimizing the memory performance in edge devices has become an important research focus. In this work, we note (based on some preliminary quantitative studies) that the memory requirements of such application-specific systems tend to differ drastically from traditional general-purpose computing systems. Inspired by these findings and also through drawing inspiration from the human brain (which excels at being highly adaptive), we design a digital memory framework that can continually adapt to the specific needs of different edge devices. This adaption is made possible through a continual reinforcement-based learning methodology, and it aims at creating a digital memory framework that is always self-aware of the data it hold and queries being made. Through a methodical implementation of the framework, we demonstrate its effectiveness for different use-cases, settings, and hyperparameters in comparison with traditional content-addressable memory. © 2020 IEEE.

关键词： Internet of things

来源：评论

学校读者我要写书评

暂无评论

Robotic FDM for free-form fabrication: evaluating adaptive non-planar slicing with different contour methods

引用

Rapid Prototyping Journal 2025年第11期31卷 160-178页

作者： Furlan, Valentina Insero, Federico Giberti, Hermes Department of Electrical Computer and Biomedical Engineering Università di Pavia Pavia Italy

Purpose: This study aims to investigate and compare three nonplanar (NP) slicing algorithms. The algorithms aim to control the layer thickness variation (LTV), which is a common issue in supportless fabrication of free-form parts. The comparison underlines the differences between theoretical and real scenarios, resulting in guidelines for toolpath generation of complex objects. Design/methodology/approach: The algorithm comparison uses a representative complex geometry, i.e. the quarter of torus. It presents an increasing overhang and constant curvature. It is represented using a parametric definition in the first two algorithms and by triangular meshes as the real scenario. The algorithms work on the contouring stage, whereas a B-spline approach is used to build the inner side of layers. Constant layer thickness (LT) is imposed, and an adaptive approach is adopted to avoid over-extrusion. Three algorithms are validated with a robotized fused deposition modeling system. LTs are measured on cross-sectioned samples and compared with the theoretical cases. Findings: The results show that two algorithms can provide an LTV of about 0% on the contour. Nevertheless, the theoretical results on the inner side are divergent from the previous evidence, moving on to higher LTV (approximately 90%). The need for an adaptive approach is demonstrated, resulting in an LTV reduction (approximately 30%). Printed parts present the same trends of theoretical results confirming the algorithms’ capabilities. Originality/value: The work shows, for the first time, a comparison between NP slicing techniques. The LTV problem in hollow and filled components is analyzed through theoretical and experimental evidence. The results are promising for supportless fabrication of free-form parts. © 2025, Valentina Furlan, Federico Insero and Hermes Giberti.

关键词： Fabrication

来源：评论

学校读者我要写书评

暂无评论

TPUXtract: An Exhaustive Hyperparameter Extraction Framework

IACR Transactions on Cryptographic Hardware and Embedded Sys...

引用

IACR Transactions on Cryptographic Hardware and Embedded Systems 2025年第1期2025卷 78-103页

作者： Kurian, Ashley Dubey, Anuj Yaman, Ferhat Aysu, Aydin Department of Electrical and Computer Engineering North Carolina State University United States

Model stealing attacks on AI/ML devices undermine intellectual property rights, compromise the competitive advantage of the original model developers, and potentially expose sensitive data embedded in the model’s behavior to unauthorized parties. While previous research works have demonstrated successful side-channelbased model recovery in embedded microcontrollers and FPGA-based accelerators, the exploration of attacks on commercial ML accelerators remains largely unexplored. Moreover, prior side-channel attacks fail when they encounter previously unknown models. This paper demonstrates the first successful model extraction attack on the Google Edge Tensor Processing Unit (TPU), an off-the-shelf ML accelerator. Specifically, we show a hyperparameter stealing attack that can extract al l layer configurations including the layer type, number of nodes, kernel/filter sizes, number of filters, strides, padding, and activation function. Most notably, our attack is the first comprehensive attack that can extract previously unseen models. This is achieved through an online template-building approach instead of a pre-trained ML-based approach used in prior works. Our results on a black-box Google Edge TPU evaluation show that, through obtained electromagnetic traces, our proposed framework can achieve 99.91% accuracy, making it the most accurate one to date. Our findings indicate that attackers can successfully extract various types of models on a black-box commercial TPU with utmost detail and call for countermeasures. © 2025, Ruhr-University of Bochum. All rights reserved.

关键词： Network security

来源：评论

学校读者我要写书评

暂无评论

Innovative Design for Mutual Coupling Reduction in Dual-Element Array Antennas for ISM Applications Using Whale Optimization Algorithm

引用

Progress in Electromagnetics Research C 2025年 152卷 121-129页

作者： Atashpanjeh, Elham Rezaei, Pejman Department of Electrical and Computer Engineering Semnan University Semnan35131-19111 Iran

This paper introduces a decoupled dual-element array antenna designed to address the challenges of mutual coupling between elements. To tackle this issue, a neutralization line is strategically incorporated to suppress leaky surface currents, while ensuring that the antenna’s central frequency and radiation pattern remain intact. The dimensions of the neutralization line are carefully optimized using the Whale Optimization Algorithm (WOA) to achieve the best possible performance, focusing on minimizing mutual coupling and enhancing gain. By placing the neutralization line nearby between the two elements, surface currents are efficiently redirected back to the radiating element, preventing leakage to neighboring elements. This approach also results in a more compact structure. The proposed antenna, with overall dimensions of 50 mm × 30 mm × 1.6 mm, is simulated using analytic software. It achieves an impressive 27 dB reduction in mutual coupling and delivers an ultra-wide bandwidth of 1.2 GHz within the Industrial Scientific and Medical (ISM) band at an operating frequency of 2.4 GHz, with a measured maximum gain of −5.19 dB. The structure was fabricated, and experimental results closely matched the simulations, confirming the design’s effectiveness. By leveraging the WOA optimization method, the geometry of the neutralization line was fine-tuned to maximize performance, significantly improving inter-element decoupling. This design approach is simple yet effective and can be readily extended to other antenna array configurations, demonstrating strong potential for compact and efficient ISM band applications. © 2025, Electromagnetics Academy. All rights reserved.

关键词： Directional patterns (antenna)

来源：评论

学校读者我要写书评

暂无评论

DenseRSE-ASPPNet: An Enhanced DenseNet169 with Residual Dense Blocks and CE-HSOA-Based Optimization for IoT Botnet Detection

引用

International Journal of Advanced computer Science and Applications 2025年第4期16卷 688-699页

作者： Khan, Mohd Abdul Rahim Department of Electrical Engineering and Computer Science A’sharqiyah University IBRA-400 Oman

The growing prevalence of Internet of Things (IoT) devices has heightened vulnerabilities to botnet-based cyberattacks, necessitating robust detection mechanisms. This paper proposes DenseRSE-ASPPNet, an advanced deep learning framework for botnet detection, incorporating comprehensive preprocessing, feature extraction, and optimization. The preprocessing pipeline includes data cleaning and Min-Max normalization to ensure high-quality input data. The DenseNet169 backbone is enhanced with Residual Squeeze-and-Excitation (RSE) blocks for channel-wise attention recalibration and Atrous Spatial Pyramid Pooling (ASPP) for capturing multi-scale spatial patterns, enabling effective feature extraction. Hyperparameter optimization is performed using the Cyclone-Enhanced Humboldt Squid Optimization Algorithm (CE-HSOA), which balances global exploration and local exploitation, ensuring faster convergence and enhanced robustness. Experimental results demonstrate the superior performance of the proposed framework, achieving 99.00 per cent accuracy, 96.40 per cent sensitivity, and 99.95 per cent specificity, significantly minimizing false positives and false negatives. The proposed DenseRSE-ASPPNet provides an efficient, scalable, and effective solution for mitigating botnet threats in IoT environments. © (2025), (Science and Information Organization). All rights reserved.

关键词： Feature Selection

来源：评论

学校读者我要写书评

暂无评论

COLT: Cyclic Overlapping Lottery Tickets for Faster Pruning of Convolutional Neural Networks

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2025年第6期6卷 1664-1678页

作者： Hossain, Md. Ismail Rakib, Mohammed Elahi, M. M. Lutfe Mohammed, Nabeel Rahman, Shafin North South University Department of Electrical and Computer Engineering Dhaka1229 Bangladesh

Pruning refers to the elimination of trivial weights from neural networks. The sub-networks within an overparameterized model produced after pruning are often called Lottery tickets. This research aims to generate winning lottery tickets from a set of lottery tickets that can achieve accuracy similar to that of the original unpruned network. We introduce a novel winning ticket called Cyclic Overlapping Lottery Ticket (COLT) by data splitting and cyclic retraining of the pruned network from scratch. We apply a cyclic pruning algorithm that keeps only the overlapping weights of different pruned models trained on different data segments. Our results demonstrate that COLT can achieve similar accuracies (obtained by the unpruned model) while maintaining high sparsities. Based on object recognition and detection tasks, we show that the accuracy of COLT is on par with the winning tickets of the Lottery Ticket Hypothesis (LTH) and, at times, is better. Moreover, COLTs can be generated using fewer iterations than tickets generated by the popular Iterative Magnitude Pruning (IMP) method. In addition, we also notice that COLTs generated on large datasets can be transferred to small ones without compromising performance, demonstrating its generalizing capability. We conduct all our experiments on Cifar-10, Cifar-100, TinyImageNet, and ImageNet datasets and report superior performance than the state-of-the-art methods. The codes are available at: https://***/ismail31416/COLT. © 2020 IEEE.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

An end-to-end network for co-saliency detection in one single image

引用

Science China(Information Sciences) 2023年第11期66卷 5-22页

作者： Yuanhao YUE Qin ZOU Hongkai YU Qian WANG Zhongyuan WANG Song WANG School of Cyber Science and Engineering Wuhan University School of Computer Science Wuhan University Department of Electrical Engineering and Computer Science Cleveland State University Department of Computer Science and Engineering University of South Carolina

Co-saliency detection within a single image is a common vision problem that has not yet been well addressed. Existing methods often used a bottom-up strategy to infer co-saliency in an image in which salient regions are firstly detected using visual primitives such as color and shape and then grouped and merged into a co-saliency map. However, co-saliency is intrinsically perceived complexly with bottom-up and top-down strategies combined in human vision. To address this problem, this study proposes a novel end-toend trainable network comprising a backbone net and two branch nets. The backbone net uses ground-truth masks as top-down guidance for saliency prediction, whereas the two branch nets construct triplet proposals for regional feature mapping and clustering, which drives the network to be bottom-up sensitive to co-salient regions. We construct a new dataset of 2019 natural images with co-saliency in each image to evaluate the proposed method. Experimental results show that the proposed method achieves state-of-the-art accuracy with a running speed of 28 fps.

关键词： saliency detection convolutional neural network regional feature mapping co-saliency detection deep learning

来源：评论

学校读者我要写书评

暂无评论

Security, Latency, and Throughput of Proof-of-Work Nakamoto Consensus

引用

IEEE Transactions on Information Theory 2025年第6期71卷 4708-4731页

作者： Cao, Shu-Jie Guo, Dongning Northwestern University Department of Electrical and Computer Engineering EvanstonIL United States

This paper investigates the fundamental trade-offs between block safety, confirmation latency, and transaction throughput of proof-of-work (PoW) longest-chain fork-choice protocols, also known as PoW Nakamoto consensus. New upper and lower bounds are derived for the probability of block safety violations as a function of honest and adversarial mining rates, a block propagation delay limit, and confirmation latency measured in both time and block depth. The results include the first non-trivial closed-form finite-latency bound applicable across all delays and mining rates up to the ultimate fault tolerance. Notably, the gap between these upper and lower bounds is narrower than previously established bounds for a wide range of parameters relevant to Bitcoin and its derivatives, including Litecoin and Dogecoin, as well as Ethereum Classic. Additionally, the study uncovers a fundamental trade-off between transaction throughput and confirmation latency, ultimately determined by the desired fault tolerance and the rate at which block propagation delay increases with block size. © 1963-2012 IEEE.

关键词： Bitcoin

来源：评论

学校读者我要写书评

暂无评论

Learning and Optimization over Robust Networked Systems

引用

Performance Evaluation Review 2025年第3期52卷 23-26页

作者： Lin, I-Cheng Department of Electrical and Computer Engineering Carnegie Mellon University United States

来源：评论

学校读者我要写书评

暂无评论

Design of a Level-3 electric vehicle charging station using a 1-MW solar system via the distributed maximum power point tracking technique

引用

Clean Energy 2024年第1期8卷 23-35页

作者： Afshin Balal Michael Giesselmann Department of Electrical and Computer Engineering Texas Tech UniversityLubbockTX 79409USA

Solar power is mostly influenced by solar irradiation,weather conditions,solar array mismatches and partial shading ***,before installing solar arrays,it is necessary to simulate and determine the possible power *** power point tracking is needed in order to make sure that,at any time,the maximum power will be extracted from the photovoltaic ***,maximum power point tracking is not a suitable solution for mismatches and partial shading *** overcome the drawbacks of maximum power point tracking due to mismatches and shadows,distributed maximum power point tracking is util-ized in this *** solar farm can be distributed in different ways,including one DC-DC converter per group of modules or per *** this paper,distributed maximum power point tracking per module is implemented,which has the highest *** technology is applied to electric vehicles(EVs)that can be charged with a Level 3 charging station in<1 ***,the problem is that charging an EV in<1 hour puts a lot of stress on the power grid,and there is not always enough peak power reserve in the existing power grid to charge EVs at that ***,a Level 3(fast DC)EV charging station using a solar farm by implementing distributed maximum power point tracking is utilized to address this ***,the simulation result is reported using MATLAB®,LTSPICE and the System Advisor *** results show that the proposed 1-MW solar system will provide 5 MWh of power each day,which is enough to fully charge~120 EVs each ***,the use of the proposed photovoltaic system benefits the environment by removing a huge amount of greenhouse gases and hazardous *** example,instead of supplying EVs with power from coal-fired power plants,1989 pounds of CO_(2) will be eliminated from the air per hour.

关键词： electric vehicles(EVs) distributed maximum power point tracking solar farm DC fast-charging station step-up DC-DC converter System Advisor Model economic analysis

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：