检索结果-内蒙古大学图书馆

Error Mitigation for TDoA UWB Indoor Localization Using Unsupervised Machine Learning

IEEE SENSORS JOURNAL 2025年第1期25卷 1959-1968页

作者： Duong, Phuong Bich Van Herbruggen, Ben Broering, Arne Shahid, Adnan De Poorter, Eli Siemens AG D-85748 Garching Germany Univ Ghent Dept Engn & Architecture B-9000 Ghent Belgium Univ Ghent IDLab B-9052 Ghent Belgium Imec B-9052 Ghent Belgium

Indoor positioning systems based on ultrawideband (UWB) technology are gaining recognition for their ability to provide cm-level localization accuracy. However, these systems often encounter challenges caused by dense multipath fading, leading to positioning errors. To address this issue, in this article, we propose a novel methodology for unsupervised anchor node selection using deep embedded clustering (DEC). Our method uses an autoencoder (AE) before clustering, thereby better separating UWB features into separable clusters of UWB input signals. Afterward, we rank these clusters based on their cluster quality, allowing us to remove untrustworthy signals. Our method is novel, as it is the first error mitigation approach for time difference of arrival (TDoA)-based UWB localization that uses unsupervised machine learning (ML), thereby avoiding costly labeling efforts and significantly reducing the localization error. Our experiments show that our method can reduce the mean absolute error (MAE) by a significant 23.1% overall, and in dense multipath areas by 26.6%, and the 95th percentile error by 49.3% when compared with without anchor selection.

关键词： clustering algorithms Prevention and mitigation Location awareness Accuracy Statistical analysis Labeling Channel impulse response Machine learning Feature extraction Error correction Deep embedded clustering (DEC) indoor positioning time difference of arrival (TDoA) ultrawideband (UWB) unsupervised machine learning (ML)

来源：评论

学校读者我要写书评

暂无评论

Anomaly Detection and Business Process Orchestration for Low-Code Platform in Power System Based on Deep-Cross Model With Data Balancing

引用

IEEE ACCESS 2025年 13卷 6350-6361页

作者： Huang, Jin Xu, Ming Sheng Zhu, Chang Ming Wang, Guo An Jiangsu Elect Power Informat Technol CO Ltd Nanjing 210003 Jiangsu Peoples R China

This paper presents an innovative approach to anomaly detection in electric vehicle (EV) charging platforms, centered around four key innovations that significantly advance the field of charging infrastructure security. First, we introduce a statistically reliable data balancing methodology that addresses the critical challenge of anomaly sample scarcity, achieving a 47% improvement in detection accuracy compared to conventional sampling techniques. Second, we propose xDeepCIN, a novel deep learning architecture that uniquely combines Compressed Interaction Networks with low-order crossing structures, demonstrating a 28% reduction in feature sparsity compared to existing deep cross models. Third, we implement a hierarchical feature interaction mechanism that enables both explicit and implicit learning of high-order patterns, resulting in a 35% improvement in anomaly pattern recognition over traditional deep learning approaches. Fourth, we develop an adaptive processing framework for asynchronous heterogeneous data streams, reducing detection latency by 62% compared to current synchronous methods. Unlike prior studies that primarily focus on single-aspect improvements in either data processing or model architecture, our comprehensive approach simultaneously addresses data imbalance, feature sparsity, and processing efficiency. Experimental results across multiple real-world datasets demonstrate the framework's superior performance, achieving an AUC of 0.93 and F1-score of 0.97, representing substantial improvements over existing methods. This advancement has significant implications for the growing EV market, particularly in China, where our framework could prevent an estimated 85% of serious charging infrastructure failures.

关键词： Vectors Anomaly detection Data models Mathematical models clustering algorithms Surges Security Adaptation models Voltage fluctuations Tensors Third-party charging infrastructure low-code platform anomaly detection data balancing deep cross model

来源：评论

学校读者我要写书评

暂无评论

A Method for Abnormality Detection of Lithium-Ion Battery Packs Based on Kullback-Leibler Test and Greenwald-Khanna clustering 11th

A Method for Abnormality Detection of Lithium-Ion Battery Pa...

引用

11th Frontier Academic Forum of Electrical Engineering, FAFEE 2024

作者： Wang, Chong Liu, Yajie Song, Yuanming Wang, Yu College of System Engineering National University of Defense Technology Changsha410073 China Hunan Key Laboratory of Multi-Energy System Intelligent Interconnection Technology Changsha410073 China

ISBN: (纸本)9789819788231

During the operation of lithium-ion battery packs, there often exhibit certain abnormalities due to cell faults such as internal short circuit or unavoidable inconsistencies among cells, which affects the operation safety of the battery packs. Therefore, it is of great significance to carry out abnormality detection for the battery packs. Accordingly, this paper proposes a feature selection method based on Kullback-Leibler (K-L) test and an improved Greenwald-Khanna (GK) clustering algorithm. Initially, faults are introduced into the battery pack using the Simulink simulation platform to obtain simulation data. Subsequently, parameter identification is identified based on the Rint equivalent circuit model and the Recursive Least Squares (RLS) algorithm. Finally, better consistency features are selected using the K-L test as inputs to the GK clustering algorithm, and the K-Means++ method is used to optimize the GK algorithm for clustering features. The results demonstrate that this method can effectively distinguish between normal and abnormal batteries, achieving abnormality detection in battery packs. © Beijing Paike Culture Commu. Co., Ltd. 2025.

关键词： clustering algorithms

来源：评论

学校读者我要写书评

暂无评论

TCMS: A Multi-Sequence Log Parsing Method Based on Token Conversion

引用

IEEE TRANSACTIONS ON DEPENDABLE AND SECURE COMPUTING 2025年第3期22卷 3028-3045页

作者： Wei, Mingkuan Wen, Jigang He, Shiming Xie, Kun Liang, Wei Xie, Gaogang Li, Kenli Zhu, Ziyu Hunan Univ Coll Comp Sci & Elect Engn Key Lab Fus Comp Supercomp & Artificial Intelligen Minist Educ Changsha 410012 Peoples R China Hunan Univ Sci & Technol Sch Comp Sci & Engn Xiangtan 411201 Peoples R China Changsha Univ Sci & Technol Sch Comp & Commun Engn Hunan Prov Key Lab Intelligent Proc Big Data Trans Changsha 410114 Peoples R China Chinese Acad Sci Comp Network Informat Ctr CNIC Beijing 100045 Peoples R China Columbia Univ Elect Engn New York NY 10027 USA

Detailed system operations are recorded in logs. To ensure system reliability, developers can detect system anomalies through log anomaly detection. Log parsing, which converts semi-structured log messages into structured data, is a crucial step in log anomaly detection and advanced program analysis and verification. Despite the availability of various log parsing tools, they generally suffer from low parsing accuracy and slow efficiency due to the ignorance of variable characteristics and the use of costly pairwise comparison methods. In this paper, we propose a TCMS framework to parse logs, consisting of two main technologies. First, by studying 16 public log datasets, we find that most log variable tokens are structured variable tokens. Based on this discovery, we propose a token conversion algorithm to improve parsing accuracy. This algorithm converts the changed parts in structured variable tokens into wildcards ('< *>'), preventing these tokens from being directly identified as constant tokens. Second, to improve efficiency, we propose the LogMLCS algorithm, which intelligently constructs a graph to facilitate the extraction of common parts from multiple log messages at once, instead of using pairwise comparisons. Comprehensive experiments conducted on 16 log datasets reveal that our TCMS outperforms seven other parsing methods, achieving the highest parsing accuracy at the fastest speed. Furthermore, experimental results from running a log anomaly detection algorithm in conjunction with different log parsing methods demonstrate that TCMS significantly boosts detection accuracy. For instance, on the OpenStack dataset, our TCMS-facilitated log anomaly detection algorithm achieves a perfect F1-score, precision, and recall of 100% each, surpassing the best peer method by 32.2, 0.8, and 19.5 percentage points, respectively.

关键词： Accuracy Anomaly detection clustering algorithms Peer-to-peer computing Electronic mail Source coding Particle separators Codes Tokenization Itemsets Log parsing log analysis token conversion LogMLCS

来源：评论

学校读者我要写书评

暂无评论

New Evaluation Method for Fuzzy Cluster Validity Indices

引用

IEEE ACCESS 2025年 13卷 22728-22744页

作者： Perez-Sanchez, Ismay Medina-Perez, Miguel Angel Monroy, Raul Loyola-Gonzalez, Octavio Gutierrez-Rodriguez, Andres Eduardo Tecnol Monterrey Sch Engn & Sci Atizapan De Zaragoza 52926 Estado De Mexic Mexico Kayak Analyt Dubai U Arab Emirates NTT DATA Madrid 28050 Spain Tecnol Monterrey Inst Future Educ Monterrey 64849 Mexico MAHLE Shared Serv Monterrey 64650 Mexico

Cluster analysis is the process of associating data objects with classes of similar objects. One key aspect of the clustering problem is to determine the optimal number of groups on a data set to effectively partition it. The process of establishing accurate indicators is called cluster validity, and it evaluates the quality of a clustering, including the optimal number of clusters. A cluster validity index (CVI) is a function that carries out that evaluation based on a series of features like intra-class compactness, inter-class separation, object density, and membership degree of objects to each cluster, among others. Best-c validation protocol has been used to evaluate CVIs previously, but it makes a strong assumption about the correctness of the clustering performed. We propose a new protocol based on mapping the internal CVI values to an external CVI and performing statistical tests to analyze the results. The experiments are conducted evaluating the partitions generated by the fuzzy c-means clustering algorithm over 84 UCI datasets with 31 fuzzy CVIs. Experimental results evidenced the need for bigger experimental setups compared to previous studies. Also, they show that among the CVIs involved, there is none with a significant statistical difference over the other given by a Friedman test. However, the Wilcoxon signed-rank test demonstrates that in 1 vs. 1, the CVI KPBM outperforms the others in most of the fuzzifiers (m). Just except for m = 2.0 which consolidates the results in previous studies where none of the CVIs were able to improve consistently among the selected group of CVIs.

关键词： clustering algorithms Partitioning algorithms Indexes Protocols Prediction algorithms Hands Weather forecasting Technological innovation Organizations Object recognition Fuzzy cluster validation index evaluation method fuzzy clustering fuzzy sets

来源：评论

学校读者我要写书评

暂无评论

Learning From Crowds by Class-Specific Instance Weighting

引用

IEEE TRANSACTIONS ON EMERGING TOPICS IN COMPUTATIONAL INTELLIGENCE 2025年

作者： Ji, Qiang Jiang, Liangxiao Zhang, Wenjun Li, Chaoqun China Univ Geosci Sch Comp Sci Wuhan 430074 Peoples R China China Univ Geosci Sch Math & Phys Wuhan 430074 Peoples R China

In crowdsourcing scenarios, we can acquire each instance's multiple noisy label set from crowd workers and then infer its integrated label via label integration. To further enhance the quality of integrated labels, some noise correction algorithms have been proposed in recent years. Most of them aim to partition the dataset into a clean set and a noise set, followed by training one or multiple classifiers on the clean set to rectify the instances in the noise set. However, they overlook the fact that the class distribution of the clean set is often inconsistent with that of the noise set, resulting in the subpar correction performance of trained classifiers. To mitigate this inconsistency, this paper proposes a class-specific instance weighting-based noise correction (CIWNC) algorithm. In CIWNC, each class's weight is computed based on the class distribution of the clean set firstly. Subsequently, a class-specific weight is computed for each instance using the weight of the class that its integrated label belongs to, as well as its multiple noisy label set. Finally, a classifier is trained on the instance weighted clean set to rectify the instances in the noise set. Experimental results on 34 simulated and two real-world datasets demonstrate that CIWNC outperforms existing state-of-the-art noise correction algorithms significantly.

关键词： Noise Classification algorithms Noise measurement Crowdsourcing Prediction algorithms clustering algorithms Supervised learning Training Inference algorithms Partitioning algorithms Crowdsourcing learning noise correction instance weighting class-specific instance weighting

来源：评论

学校读者我要写书评

暂无评论

A Collaborative Programmable LFA Defense Using Temporal Graph Learning in AIoT

引用

IEEE INTERNET OF THINGS JOURNAL 2025年第3期12卷 2514-2529页

作者： Li, Yikun Liu, Ying Xia, Yu Zhang, Weiting Quan, Wei Kang, Jiawen Zhang, Hongke Beijing Jiaotong Univ Sch Elect & Informat Engn Beijing 100044 Peoples R China Peng Cheng Lab Dept New Networks Shenzhen 518000 Guangdong Peoples R China Guangdong Univ Technol Sch Automat Guangzhou 510006 Peoples R China Guangdong Univ Technol Guangdong Key Lab IoT Informat Technol Guangzhou 510006 Peoples R China

In the current era of rapid advancements in Artificial Intelligence of Things (AIoT), with the increase in cloud data center operations and the limited security computing capabilities of AIoT terminal devices, link flooding attack (LFA) has emerged as a complex and stealthy new threat. However, the existing defense methods based on programmable networks usually have issues of slow offline inference and delayed defense activation. To address these issues, we propose a collaborative programmable defense framework (CPDTG) to predict, detect, and mitigate LFA. First, an early attack intention prediction model based on temporal graph learning (TGL) is proposed to accurately locate attacks and promptly activate defenses to save resource consumption during idle time. Second, a switch-native clustering algorithm independent of the global perspective is introduced for line-speed detection of LFA. The unsupervised algorithm does not rely on labeled datasets for training, which enhances its robustness against differentiated attack scenarios. Third, we propose a distributed defense mechanism that achieves the pushback deployment of adaptive rate-limiting strategies. Compressing the potential attack vector space effectively increases the difficulty of launching rolling attacks. Extensive experimental validation demonstrates the effectiveness of the proposed CPDTG in predicting and defending against LFA.

关键词： Switches Servers Cloud computing Internet of Things Prevention and mitigation Predictive models Denial-of-service attack Computer crime Collaboration clustering algorithms Attack intention prediction link flooding attack (LFA) programmable data plane (PDP) temporal graph learning (TGL)

来源：评论

学校读者我要写书评

暂无评论

TIRDH: A Novel Three-Shadow-Image Reversible Data Hiding Algorithm Using Weight and Modulo

引用

IEEE ACCESS 2025年 13卷 49232-49248页

作者： Chin, Li-Chian Li, Yu-Chen Hsieh, Hung-Mo Wang, Chung-Ming Natl Chung Hsing Univ Dept Comp Sci & Engn Taichung 40223 Taiwan Natl Taiwan Univ Dept Comp Sci & Informat Engn Taipei 106319 Taiwan

Reversible data hiding (RDH) in image media is crucial for securely embedding confidential information while ensuring complete recovery of both the cover image and the hidden data. Traditional RDH techniques often struggle to balance embedding capacity and image fidelity, limiting their practical effectiveness. This paper introduces TIRDH, a novel three-shadow-image RDH scheme that employs weight and modulo operations to achieve high-capacity embedding while preserving image quality. The method embeds an M-ary secret message within a shadow pixel cluster using a 3-tuple weight vector, modulo operations, and a variation table to guide pixel modifications. This guarantees data integrity without pixel underflow or overflow, thereby avoiding the falling-off boundary problem (FOBP) and ensuring reversibility. Additionally, the scheme offers flexible embedding rates, adjustable through a single parameter and weight vector, making it highly adaptable to various user requirements, image types, and application scenarios. Experimental results from three standard image databases demonstrate its effectiveness, achieving embedding rates ranging from 0.04 to 1.6814 bits per pixel while maintaining high image quality, with an average PSNR ranging from 47.65 to 62.65 dB. Furthermore, its prediction mechanism attains 98.2% accuracy, enabling precise performance estimation before actual message embedding. This predictive capability enhances efficiency by allowing users to anticipate the impact on image quality beforehand. Security assessments confirm strong resilience against RS steganalysis and pixel difference histogram attacks, providing enhanced protection for confidential data. Comparisons with 12 state-of-the-art methods show that the proposed approach outperforms competitors when embedding fewer than 50,000 secret bits and surpasses another set of 12 schemes when the embedding rate exceeds 1.0 bits per pixel. For 1,000 test images with an embedding rate of 1.4 bits per pixel,

关键词： Image quality Security Prediction algorithms Vectors Standards Media Robustness Histograms clustering algorithms Indexes Reversible data hiding three-shadow-image weights modulus operator prediction embedding rate embedding capacity image quality

来源：评论

学校读者我要写书评

暂无评论

ASDroid: Resisting Evolving Android Malware With API Clusters Derived From Source Code

引用

IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY 2025年 20卷 1822-1835页

作者： Hu, Qihua Wang, Weiping Song, Hong Guo, Song Zhang, Jian Zhang, Shigeng Cent South Univ Sch Comp Sci & Engn Changsha 410083 Peoples R China Hong Kong Univ Sci & Technol Dept Comp Sci & Engn Hong Kong Peoples R China

Machine learning-based Android malware detection has consistently demonstrated superior results. However, with the continual evolution of the Android framework, the efficacy of the deployed models declines markedly. Existing solutions necessitate frequent and expensive model retraining to resist the constant evolution of malware accompanying Android framework updates. To address this, we introduce a solution called ASDroid, which generalizes specific APIs into similar API clusters to counteract evolving Android malware threats. One primary challenge lies in identifying analogous API clusters that correspond to specific APIs. Our approach involves extracting semantic information from open-source API source code to construct a heterogeneous information graph, and utilizing embedding algorithms to obtain semantic vector representations of APIs. APIs that are close in embedding distance are presumed to have similar semantics. Our dataset encompasses Android applications spanning nine years from 2011 to 2019. In comparison to existing Android malware detection model aging mitigation solutions like APIGraph, SDAC and MaMaDroid, ASDroid demonstrates greater accuracy and more effective at resisting continuously evolving malware.

关键词： Source coding Prevention and mitigation Semantics clustering algorithms Resists Aging Malware Vectors Data mining Resilience Mobile security android malware detection model aging

来源：评论

学校读者我要写书评

暂无评论

Three-in-One: Robust Enhanced Universal Transferable Anti-Facial Retrieval in Online Social Networks

引用

IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY 2025年 20卷 4408-4421页

作者： Lv, Yunna Tang, Long Ye, Dengpan Xie, Caiyun Deng, Jiacheng He, Yiheng Shen, Sipeng Wuhan Univ Sch Cyber Sci & Engn Key Lab Aerosp Informat Secur & Trusted Comp Minist Educ Wuhan 430072 Peoples R China

Deep hash-based retrieval techniques are widely used in facial retrieval systems to improve the efficiency of facial matching. However, it also carries the danger of exposing private information. Deep hash models are easily influenced by adversarial examples, which can be leveraged to protect private images from malicious retrieval. The existing adversarial example methods against deep hash models focus on universality and transferability, lacking the research on its robustness in online social networks (OSNs), which leads to their failure in anti-retrieval after post-processing. Therefore, we provide the first in-depth discussion on robustness in universal transferable anti-facial retrieval and propose Three-in-One Adversarial Perturbation (TOAP). Specifically, we construct a local and global Compression Generator (CG) to simulate complex post-processing scenarios, which can be used to mitigate perturbation. Then, we propose robust optimization objectives based on the discovery of the variation patterns of model's distribution after post-processing, and generate adversarial examples using these objectives and meta-learning. Finally, we iteratively optimize perturbation by alternately generating adversarial examples and fine-tuning the CG, balancing the performance of perturbation while enhancing CG's ability to mitigate them. Numerous experiments demonstrate that, in addition to its advantages in universality and transferability, TOAP significantly outperforms current state-of-the-art methods in multiple robustness metrics. It further improves universality and transferability by 5% to 28%, and achieves up to about 33% significant improvement in several simulated post-processing scenarios as well as mainstream OSNs, demonstrating that TOAP can effectively protect private images from malicious retrieval in real-world scenarios.

关键词： Perturbation methods Robustness Privacy Image retrieval Training Optimization clustering algorithms Social networking (online) Transform coding Image coding Deep hash facial privacy protection robust adversarial perturbation online social networks

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：