检索结果-内蒙古大学图书馆

Enhancing Passive Wi-Fi Sensing-Based Crowd Analysis via a Deep Compressed Sensing Approach

IEEE INTERNET OF THINGS JOURNAL 2025年第8期12卷 10222-10237页

作者： Chang, Wenbo Huang, Baoqi Gao, Yuanda Hao, Lifei Jia, Bing Inner Mongolia Univ Coll Comp Sci Engn Res Ctr Ecol Big Data Minist Educ Hohhot 010021 Peoples R China Inner Mongolia Univ Coll Comp Sci Inner Mongolia Key Lab Wireless Networking & Mobil Hohhot 010021 Peoples R China Inner Mongolia Elect Technol Co Ltd Software Dev & Res Dept Ordos 017000 Peoples R China Univ Michigan Ann Arbor Sch Engn Ann Arbor MI 48109 USA

Real-time awareness of crowd counts and distribution is vital for applications like crowd management, traffic control, and urban planning. Compared to vision-based methods, passive Wi-Fi sensing-based approaches offer advantages, such as lower deployment costs, broader coverage, and greater scalability, and thus have become prevalent for fine-grained crowd analysis, e.g., estimating mobile device locations and subsequently inferring crowd density maps (CDMs). Therein, substantial localization errors are often incurred due to significant measurement noises, and even worse, the sporadic active scanning introduces sparsity in localization-based density maps (DMs), degrading the accuracy of resulting crowd analysis. To address these challenges, Wi-FDM, a passive Wi-Fi sensing-based framework for fine-grained CDM estimation, is proposed. Specifically, to mitigate the considerable localization errors, two models based on convolutional neural networks are employed to capture the relationship between received signal strength fingerprints and various locations more effectively;to alleviate the sparsity in localization-based DMs, deep compressed sensing networks are developed to reconstruct fine-grained CDMs by recover missing spatial details;to efficiently transfer the CDM reconstruction capabilities developed on semi-synthetic datasets to diverse real-world scenarios, learnable linear and convolutional residual blocks were designed to establish correlations across environments, addressing variations in scene characteristics and improving model generalization. Experimental results on real-world datasets demonstrate Wi-FDM not only surpasses state-of-the-art methods by 41.8% and 29.3% in crowd counts and distribution estimation but is also applicable to both indoor and outdoor scenes, demonstrating its potential for crowd management and analysis.

关键词： Wireless fidelity Sensors Location awareness Mobile handsets Accuracy Fingerprint recognition Image reconstruction Estimation Probes Pedestrians Compressed sensing (CS) convolutional neural network (CNN) crowd analysis passive Wi-Fi sensing Transformer

来源：评论

学校读者我要写书评

暂无评论

SAHSD: Enhancing Hate Speech Detection in LLM-Powered Web Applications via Sentiment Analysis and Few-Shot Learning 25

SAHSD: Enhancing Hate Speech Detection in LLM-Powered Web Ap...

引用

Proceedings of the ACM on Web Conference 2025

作者： Yulong Wang Hong Li Ni Wei State Key Laboratory of Networking and Switching Technology School of Computer Science (National Pilot Software Engineering School) Beijing University of Posts and Telecommunications Beijing China College of Big Data and Information Engineering Guizhou University Guiyang Guizhou China School of Communication Science and Engineering Fudan University Shanghai China

ISBN: (纸本)9798400712746

As large language models (LLMs) increasingly power web applications, including social networks, the challenge of moderating hate speech has become a critical concern for the Web. These LLM-powered applications, while offering near-human interaction capabilities, are vulnerable to harmful or biased content due to imperfect training data scraped from the Web. Current hate speech detection methods often struggle with limited annotated data, especially for real-time moderation on these platforms. This paper introduces Sentiment-Aided Hate Speech Detection (SAHSD), a novel approach designed to enhance hate speech detection specifically in LLM-powered web applications. By treating hate speech detection as a few-shot learning task, SAHSD utilizes sentiment analysis to refine pre-trained language models (LM) for improved accuracy in recognizing harmful content. SAHSD first employs publicly available sentiment datasets to train a sentiment analysis model, which is then fine-tuned by merging sentiment prompts with hate speech prompts, enabling efficient and accurate detection even with limited training samples. The effectiveness of SAHSD is demonstrated through experiments on widely used web-sourced datasets like SBIC and HateXplain. SAHSD achieves an exceptional F1-score of 0.99 with only 64 training samples and outperforms advanced techniques such as ToKen, MRP, and HARE, with significant improvements of 33% on SBIC and 95% on HateXplain. SAHSD surpasses GPT-4 in generalization performance across multiple datasets, showing an 8% improvement when trained on equal-sized samples. These results underscore SAHSD's potential to enhance content moderation in LLM-driven web platforms, contributing to a safer, more inclusive and accountable Web ecosystem.

关键词： hate speech

来源：评论

学校读者我要写书评

暂无评论

The PathStar™ access server:: Facilitating carrier-scale packet telephony

引用

BELL LABS TECHNICAL JOURNAL 1998年第4期3卷 86-102页

作者： Fossaceca, LM Sandoz, AD Winterbottom, P Carrier Data Networking Grp PathStarTM Access Server Core Software Dev Team Murray Hill NJ USA Lucents Data Networking Grp PathStarTM Access Server Grp Red Bank NJ USA

The maturation of large-scale Internet protocol (IP) routing and the standardization of voice over IP (VoIP) protocols have accelerated the deployment of packet telephony products in the public network. Service providers such as the Regional Bell Operating Companies, competitive local exchange carriers, multiple system operators, and Internet service providers (ISPs) are beginning to seek out next-generation products to consolidate voice, video, and data onto converged packet-based networks. To address this market, Lucent Technologies has developed and introduced a suite of data networking and packet telephony products, one of which is the PathStar(TM) Access Server. This revolutionary new product integrates Bell Labs software innovations such as the inferno(R) operating system with access and interconnection technologies developed by the Switching, Access, and data networking areas of Lucent Technologies. The PathStar Access Server combines the functionality and features of traditional circuit switches with advanced packet routing techniques to enable end-to-end converged network solutions. This paper presents an overview of the PathStar Access Server and briefly explores its hardware and software architectures.

关键词： Internet telephony

来源：评论

学校读者我要写书评

暂无评论

Effective quality control during disaster recovery

引用

BELL LABS TECHNICAL JOURNAL 2004年第2期9卷 163-171页

作者： Krock, RE Member Technical Staff Switching Software Optical Networking and Data Solutions Department Lucent Worldwide Services Lisle Illinois

Since September 11, 2001, the increased awareness of homeland security issues has focused a great deal of attention on disaster recovery planning. Plans for mobilizing restoration forces and equipment along with project management procedures to be used during a recovery, are now, or should be, part of every company's business continuity plan. During a telecommunications disaster recovery, the primary emphasis is on rapid restoration of service. This is appropriate, however, the level of quality with which the work is accomplished will have a great bearing on how the site performs in the future. This paper describes the use of a small, dedicated quality team that balances the need for speed with the need to maintain high-quality standards during restoration services, using examples from the Hinsdale fire restoration of 1988. (C) 2004 Lucent Technologies Inc.

关键词： DISASTERS CRISIS management EMERGENCY management SEPTEMBER 11 Terrorist Attacks, 2001 PROJECT management

来源：评论

学校读者我要写书评

暂无评论

SEQUIN: An SNMP-based MPLS network monitoring system

引用

BELL LABS TECHNICAL JOURNAL 2003年第1期8卷 95-111页

作者： Thottan, MK Swanson, GK Cantone, M Ho, TK Ren, J Paul, S Bell Labs Crawford Hill Networking Software Res Ctr Holmdel NJ 07733 USA Bell Labs Crawford Hill Wireless Networking Res Dept Holmdel NJ 07733 USA Bell Labs Crawford Hill Data Networking Res Ctr Networking Software Res Dept Holmdel NJ 07733 USA Bell Labs Comp Sci Res Ctr Murray Hill NJ 07974 USA Bell Labs Crawford Hill Network & Serv Management Res Dept Holmdel NJ 07733 USA Bell Labs Crawford Hill Networking Software Res Holmdel NJ 07733 USA

Multiprotocol label switching (MPLS)-based networks are gaining popularity with service providers because they retain the flexibility of Internet protocol (IP) while providing support for quality of service (QoS) through differentiated services (DiffServ). The deployment of such networks requires service providers to ensure that service-level agreements (SLAs) with their customers are being met while guaranteeing cost-effective utilization of their network. The key to achieving these objectives is an efficient monitoring system that tracks QoS metrics such as bandwidth utilization, delay, and packet loss. These metrics can be tracked at multiple granularities, such as per customer, per MPLS tunnel, and per physical network pipe. SEQUIN is a simple network management protocol (SNMP)-based monitoring system capable of tracking QoS metrics in near real time. Some of the challenges addressed by SEQUIN are synthesis of end-to-end QoS using polled data from different network elements, visualization of these metrics across the network, identification of performance problems by correlating the monitored parameters, and design of scalable polling strategies to minimize management traffic. (C) 2003 Lucent Technologies Inc.

关键词： .service providers real time DiffServ bandwidth utilization QoS metrics Monitoring System service-level Multiprotocol label switching MPLS MPLS tunnel

来源：评论

学校读者我要写书评

暂无评论

Weakness of Enhanced DNA and ElGamal Cryptosystem for Secure data Storage and Retrieval in Cloud 13

Weakness of Enhanced DNA and ElGamal Cryptosystem for Secure...

引用

13th International Conference on Information Technology in Medicine and Education, ITME 2023

作者： Fan, Yibiao Liu, Chenglian Yang, Yuxia Fang, Jie Software Engineering Institute of Guangzhou Department of Networking Technology Guangzhou510990 China School of Big Data and Artificial Intelligence Fujian Polytechnic Normal University Fuqing250300 China

ISBN: (纸本)9798350319156

DNA cryptography has been developed for nearly thirty years, and there is a lot of research on DNA based encryption and decryption. In 2018, Thangavel and Varlakshmi proposed an article 'Enhanced DNA and ElGamal cryptosystem for secure data storage and retrieval in cloud', they provide two cryptosystem schemes DNA and ElGamal. In this paper the authors would like to analyze security to DNA scheme, this is our research goal. We also prove it may occur insecure when the XOR operation using in special environment. The mathematical Proof and experimental evidence be demonstrated in this article, this is our method. The results point out it is not secure to Thangavel and Varlakshmi's DNA protocol. © 2023 IEEE.

关键词： DNA

来源：评论

学校读者我要写书评

暂无评论

New Adversarial Image Detection Based on Sentiment Analysis

引用

IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS 2024年第10期35卷 14060-14074页

作者： Wang, Yulong Li, Tianxiang Li, Shenghong Yuan, Xin Ni, Wei Beijing Univ Posts & Telecommun Sch Comp Sci Natl Pilot Software Engn Sch State Key Lab Networking & Switching Technol Beijing 100876 Peoples R China Common wealth Sci & Ind Res Org CSIRO Data61 Marsfield Sydney NSW 2122 Australia

Deep Neural Networks (DNNs) are vulnerable to adversarial examples, while adversarial attack models, e.g., DeepFool, are on the rise and outrunning adversarial example detection techniques. This paper presents a new adversarial example detector that outperforms state-of-the-art detectors in identifying the latest adversarial attacks on image datasets. Specifically, we propose to use sentiment analysis for adversarial example detection, qualified by the progressively manifesting impact of an adversarial perturbation on the hidden-layer feature maps of a DNN under attack. Accordingly, we design a modularized embedding layer with the minimum learnable parameters to embed the hidden-layer feature maps into word vectors and assemble sentences ready for sentiment analysis. Extensive experiments demonstrate that the new detector consistently surpasses the state-of-the-art detection algorithms in detecting the latest attacks launched against ResNet and Inception neutral networks on the CIFAR-10, CIFAR-100 and SVHN datasets. The detector only has about 2 million parameters, and takes shorter than 4.6 milliseconds to detect an adversarial example generated by the latest attack models using a Tesla K80 GPU card.

关键词： Adversarial example detection deep learning neural network sentiment analysis

来源：评论

学校读者我要写书评

暂无评论

A secure rational quantum state sharing protocol

引用

Science China(Information Sciences) 2018年第2期61卷 204-215页

作者： Zhao DOU Gang XU Xiu-Bo CHEN Xin LIU Yi-Xian YANG Information Security Center State Key Laboratory of Networking and Switching TechnologyBeijing University of Posts and Telecommunications School of Software Engineering Beijing University of Posts and Telecommunications School of Information Engineering Inner Mongolia University of Science and Technology School of Computer Science Shaanxi Normal University Gui Zhou University Guizhou Provincial Key Laboratory of Public Big Data

A novel rational protocol to share two arbitrary qubits among multiple parties is investigated in this paper. First, the protocol is presented, which is learned from Li et al.’s protocol. Second, the utility,security, correctness, fairness, Nash equilibrium, and Pareto optimality of our scheme are discussed in detail,where the utility, correctness, and fairness of rational quantum state sharing protocols are creatively given because the agent who recovers the state plays a different and more important role. Another important point is that assumptions about our protocol are more practical and suitable than existing protocols.

关键词： rational quantum state sharing Nash equilibrium secure correct

来源：评论

学校读者我要写书评

暂无评论

Toward Smart Vehicle-to-Everything-Connected Powertrains: Driving Real Component Test Benches in a Fully Interactive Virtual Smart City

引用

IEEE VEHICULAR TECHNOLOGY MAGAZINE 2021年第1期16卷 75-82页

作者： Eisenbarth, Markus Wegener, Marius Scheer, Rene Andert, Jakob Buse, Dominik S. Klingler, Florian Sommer, Christoph Dressler, Falko Reinold, Peter Gries, Rafael Rhein Westfal TH Aachen Inst Combust Engines Jr Professorship Mechatron Syst Combust Engines Aachen Germany Software Innovat Campus Paderborn Germany TU Berlin Chair Data Commun & Networking Berlin Germany Paderborn Univ Dept Comp Sci Comp Networks Grp Paderborn Germany Paderborn Univ Comp Sci Paderborn Germany TU Berlin Sch Elect Engn & Comp Sci Data Com municat & networking Berlin Germany dSPACE GmbH Paderborn Germany Denso Automot Deutschland Electrificat Business Unit Dev High Voltage Elect Machines Wegberg Germany

In the context of increasing electrification and the automation of future mobility, research and development of efficient powertrains requires enhanced test methods. One important aspect to consider is the complex interactions among a smart driving strategy, traffic flow, an individual vehicle's energy demand and emissions, and the corresponding overall values. Until now, the respective development domains of traffic flow control, powertrain control, component design, and intervehicle communication have usually been considered separately. This article presents a methodology that combines these areas and enables developers to obtain deep and highly realistic system insights, taking into account the mutual interactions among the domains. For this purpose, an X-in-the-loop validation platform is constructed that builds upon vehicular networking, road traffic, and a vehicle simulator, all coupled with two real drive-unit component test benches. We further show how this methodology can be used to investigate the effects of a novel predictive powertrain control algorithm that takes vehicle-to-everything (V2X) communication into account. Focusing on a typical urban reference route, we demonstrate that our algorithm enables performance to be maintained using electric motors with reduced specifications.

关键词： Mechanical power transmission Road traffic Prediction algorithms Electric motors Vehicle-to-everything Couplings Real-time systems Intelligent vehicles

来源：评论

学校读者我要写书评

暂无评论

Specifying and Composing Web Services with an Environment Ontology-Based Approach

引用

INTERNATIONAL JOURNAL OF WEB SERVICES RESEARCH 2010年第3期7卷 73-92页

作者： Wang, Puwei Jin, Zhi Liu, Lin Wu, Budan Renmin Univ China Key Lab Data Engn & Knowledge Engn Minist Educ Beijing Peoples R China Peking Univ Key Lab High Confidence Software Technol Minist Educ Beijing Peoples R China Chinese Acad Sci Beijing 100864 Peoples R China Tsinghua Univ Sch Software Beijing Peoples R China Beijing Univ Posts & Telecommun State Key Lab Networking & Switching Technol Beijing Peoples R China

Precise capability specification is the key for identifying and composing the right Web services. This paper specifies service capabilities in terms of the environment entities from the application domain and the effects imposed by the Web service on these entities. An environment ontology for Web services is adopted to provide formal sharable representations of the domain-specific environment entities. A hierarchical state machine is constructed for each environment entity to describe its behaviors, and the effects imposed by a Web service are described as the state transitions traces of environment entities, which define the capability of the Web service. Web service composition that satisfies a set of requested effects is then conducted by reasoning on the effects of services. The proposed approach emphasizes the external manifestation of Web services and service composition based on the effect reasoning. An example of online travel service illustrates the proposed approach.

关键词： Capability Specification Composition Environment Ontology Precise Capability Web Services

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：