Pervasive and Mobile Computing最新文献_第2页

ALIBIS: Assessing and mitigating the risk of sensitive metadata Leakage In moBile Image Sharing ALIBIS：移动图像共享中敏感元数据泄露的评估和降低风险

IF 3.5 3区计算机科学 Q2 COMPUTER SCIENCE, INFORMATION SYSTEMS

Pervasive and Mobile Computing

Pub Date : 2026-04-01 Epub Date: 2026-01-17 DOI: 10.1016/j.pmcj.2026.102171

Tran Thanh Lam Nguyen, Barbara Carminati, Elena Ferrari

Smartphones have become necessary in modern life and can replace traditional devices like cameras. The high demand for taking and sharing photos via smartphones, especially with the explosion of social networks and instant messaging, highlights the importance of smartphones. Android, the leading smartphone operating system, has continuously improved user security and privacy over its 17 years of development (2008–2025). However, security vulnerabilities still exist because of its open-source nature. This paper introduces ALIBIS, a framework that automatically estimates the risk of leakage of sensitive data contained in EXIF metadata when users share images online by combining static analysis and Large Language Models (LLMs). ALIBIS demonstrates consistent and robust estimation capabilities, achieving an average accuracy, precision, recall, and f1 score in k-fold cross-validation (k=5) of 0.8686, 0.8902, 0.881, and 0.8854, respectively. In addition, a survey of 130 global participants, including Android app developers and end-users, revealed a significant lack of awareness about image metadata and its risks: 82.3% of participants (user role) do not delete sensitive metadata before sharing images, and 62.3% do not know how to remove metadata. Furthermore, only 1.9% of participants (developer role) proactively remove EXIF metadata during programming. We propose ExifMetadataLib, a lightweight library for easy integration with Android OS, to mitigate sensitive metadata leakage.

智能手机已经成为现代生活的必需品，可以取代相机等传统设备。通过智能手机拍摄和分享照片的高需求，尤其是随着社交网络和即时通讯的爆炸式增长，凸显了智能手机的重要性。Android作为领先的智能手机操作系统，在其17年的发展（2008-2025）中不断提高用户的安全性和隐私性。然而，由于其开源特性，安全漏洞仍然存在。本文介绍了ALIBIS框架，该框架结合静态分析和大型语言模型（Large Language Models, llm），自动估计用户在线共享图像时EXIF元数据中包含的敏感数据泄露的风险。ALIBIS具有一致性和鲁棒性的估计能力，k-fold交叉验证（k=5）的平均正确率、精密度、召回率和f1得分分别为0.8686、0.8902、0.881和0.8854。此外，一项针对130名全球参与者（包括Android应用开发者和最终用户）的调查显示，人们对图像元数据及其风险的认识严重不足：82.3%的参与者（用户角色）在共享图像之前不会删除敏感元数据，62.3%的参与者不知道如何删除元数据。此外，只有1.9%的参与者（开发人员角色）在编程期间主动删除EXIF元数据。我们提出ExifMetadataLib，一个轻量级的库，易于与Android操作系统集成，以减轻敏感的元数据泄漏。

{"title":"ALIBIS: Assessing and mitigating the risk of sensitive metadata Leakage In moBile Image Sharing","authors":"Tran Thanh Lam Nguyen, Barbara Carminati, Elena Ferrari","doi":"10.1016/j.pmcj.2026.102171","DOIUrl":"10.1016/j.pmcj.2026.102171","url":null,"abstract":"<div><div>Smartphones have become necessary in modern life and can replace traditional devices like cameras. The high demand for taking and sharing photos via smartphones, especially with the explosion of social networks and instant messaging, highlights the importance of smartphones. Android, the leading smartphone operating system, has continuously improved user security and privacy over its 17 years of development (2008–2025). However, security vulnerabilities still exist because of its open-source nature. This paper introduces ALIBIS, a framework that automatically estimates the risk of leakage of sensitive data contained in EXIF metadata when users share images online by combining static analysis and Large Language Models (LLMs). ALIBIS demonstrates consistent and robust estimation capabilities, achieving an average accuracy, precision, recall, and f1 score in k-fold cross-validation (k=5) of 0.8686, 0.8902, 0.881, and 0.8854, respectively. In addition, a survey of 130 global participants, including Android app developers and end-users, revealed a significant lack of awareness about image metadata and its risks: 82.3% of participants (user role) do not delete sensitive metadata before sharing images, and 62.3% do not know how to remove metadata. Furthermore, only 1.9% of participants (developer role) proactively remove EXIF metadata during programming. We propose ExifMetadataLib, a lightweight library for easy integration with Android OS, to mitigate sensitive metadata leakage.</div></div>","PeriodicalId":49005,"journal":{"name":"Pervasive and Mobile Computing","volume":"117 ","pages":"Article 102171"},"PeriodicalIF":3.5,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146024341","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

SMOTE-Enhanced CNN-Bi-LSTM for wearable sensor-based human activity recognition 基于smote增强的CNN-Bi-LSTM可穿戴传感器的人体活动识别

IF 3.5 3区计算机科学 Q2 COMPUTER SCIENCE, INFORMATION SYSTEMS

Pervasive and Mobile Computing

Pub Date : 2026-04-01 Epub Date: 2026-01-08 DOI: 10.1016/j.pmcj.2026.102161

Ahmed Arafa , Hadeer Harfoush , Nawal El-Fishawy , Marwa Radad

Human Activity Recognition (HAR) refers to the automatic recognition of various human physical activities such as walking, sitting, and standing. HAR based on wearable sensors and smartphones has gained significant attention due to its wide range of applications in healthcare, sports, rehabilitation, and smart environments. Despite extensive research, challenges remain in modeling complex spatial–temporal dependencies within activity sequences and addressing class imbalance issues in sensor-based datasets. In this paper, we propose a hybrid deep learning model that integrates a Convolutional Neural Network (CNN) for spatial feature extraction followed by a Bidirectional Long Short-Term Memory (Bi-LSTM) for bi-directional sequential analysis and a fully connected layer for classifying the different types of activities. To address data imbalance and enhance the model robustness, three oversampling techniques — Random Oversampling (ROS), Adaptive Synthetic Sampling (ADASYN), and Synthetic Minority Over-sampling Technique (SMOTE) — were experimentally evaluated, with SMOTE demonstrating superior performance. The proposed model was trained and evaluated on six publicly available benchmark datasets: MHealth, PAMAP2, WISDM, UCI-HAR, USC-HAD and Opportunity datasets, achieving F1-score at 100%, 97.99%, 99.0%, 94.81%, 91.13% and 90.95% respectively. Comparative results demonstrate that the proposed framework outperforms several state-of-the-art methods across multiple datasets, confirming its robustness, reliability, and generalization capability for diverse human activity recognition scenarios.

人类活动识别（Human Activity Recognition， HAR）是指对人类行走、坐、站等各种身体活动的自动识别。基于可穿戴传感器和智能手机的HAR因其在医疗保健、运动、康复和智能环境中的广泛应用而备受关注。尽管进行了广泛的研究，但在模拟活动序列中复杂的时空依赖关系和解决基于传感器的数据集中的类不平衡问题方面仍然存在挑战。在本文中，我们提出了一种混合深度学习模型，该模型集成了用于空间特征提取的卷积神经网络（CNN）、用于双向序列分析的双向长短期记忆（Bi-LSTM）和用于分类不同类型活动的完全连接层。为了解决数据不平衡和增强模型鲁棒性，实验评估了三种过采样技术——随机过采样（ROS）、自适应合成采样（ADASYN）和合成少数过采样技术（SMOTE）， SMOTE表现出优异的性能。该模型在MHealth、PAMAP2、WISDM、UCI-HAR、USC-HAD和Opportunity 6个公开的基准数据集上进行了训练和评估，f1得分分别为100%、97.99%、99.0%、94.81%、91.13%和90.95%。对比结果表明，所提出的框架在多个数据集上优于几种最先进的方法，证实了其在不同人类活动识别场景中的鲁棒性、可靠性和泛化能力。

{"title":"SMOTE-Enhanced CNN-Bi-LSTM for wearable sensor-based human activity recognition","authors":"Ahmed Arafa , Hadeer Harfoush , Nawal El-Fishawy , Marwa Radad","doi":"10.1016/j.pmcj.2026.102161","DOIUrl":"10.1016/j.pmcj.2026.102161","url":null,"abstract":"<div><div>Human Activity Recognition (HAR) refers to the automatic recognition of various human physical activities such as walking, sitting, and standing. HAR based on wearable sensors and smartphones has gained significant attention due to its wide range of applications in healthcare, sports, rehabilitation, and smart environments. Despite extensive research, challenges remain in modeling complex spatial–temporal dependencies within activity sequences and addressing class imbalance issues in sensor-based datasets. In this paper, we propose a hybrid deep learning model that integrates a Convolutional Neural Network (CNN) for spatial feature extraction followed by a Bidirectional Long Short-Term Memory (Bi-LSTM) for bi-directional sequential analysis and a fully connected layer for classifying the different types of activities. To address data imbalance and enhance the model robustness, three oversampling techniques — Random Oversampling (ROS), Adaptive Synthetic Sampling (ADASYN), and Synthetic Minority Over-sampling Technique (SMOTE) — were experimentally evaluated, with SMOTE demonstrating superior performance. The proposed model was trained and evaluated on six publicly available benchmark datasets: MHealth, PAMAP2, WISDM, UCI-HAR, USC-HAD and Opportunity datasets, achieving F1-score at 100%, 97.99%, 99.0%, 94.81%, 91.13% and 90.95% respectively. Comparative results demonstrate that the proposed framework outperforms several state-of-the-art methods across multiple datasets, confirming its robustness, reliability, and generalization capability for diverse human activity recognition scenarios.</div></div>","PeriodicalId":49005,"journal":{"name":"Pervasive and Mobile Computing","volume":"117 ","pages":"Article 102161"},"PeriodicalIF":3.5,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145915147","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

AutoCompress: Improving network efficiency for distributed wireless sensing applications AutoCompress：提高分布式无线传感应用的网络效率

IF 3.5 3区计算机科学 Q2 COMPUTER SCIENCE, INFORMATION SYSTEMS

Pervasive and Mobile Computing

Pub Date : 2026-04-01 Epub Date: 2026-01-22 DOI: 10.1016/j.pmcj.2026.102178

Yamini Shankar, Ayon Chakraborty

Radio Frequency (RF)-based wireless sensing provides a privacy-preserving and non-intrusive alternative to vision-based systems for human activity recognition and environmental monitoring. However, transmitting high dimensional Channel State Information (CSI) from multiple distributed sensors leads to severe network congestion, especially at scale. We propose AutoCompress, a model-aware framework that intelligently compresses CSI spectrograms by prioritizing the most informative elements. Using Sparse Sensor Placement Optimization for Classification (SSPOC), AutoCompress computes element-wise importance scores for each spectrogram. These scores guide a sensor-subcarrier selection strategy under bandwidth constraints, implemented via the Prioritized Weighted Subcarrier-Sensor Cover (PWSSC) algorithm. Evaluated on the UT-HAR dataset and a real-world Nexmon-based testbed, AutoCompress achieves an average

> 4000 \times

reduction in data transmission, improves network throughput by 35.5%, and reduces latency by

> 90 %

-all while maintaining high inference accuracy, compared to baseline uncompressed CSI data transmission. These results demonstrate AutoCompress as a scalable, interpretable, and bandwidth-efficient solution for distributed wireless sensing systems.

基于射频（RF）的无线传感为人类活动识别和环境监测提供了一种隐私保护和非侵入性替代基于视觉的系统。然而，从多个分布式传感器传输高维信道状态信息（CSI）会导致严重的网络拥塞，特别是在规模下。我们提出了AutoCompress，这是一个模型感知框架，通过优先考虑最具信息量的元素来智能地压缩CSI频谱图。使用稀疏传感器放置优化分类（SSPOC）， AutoCompress计算每个谱图的元素重要性分数。这些分数指导在带宽限制下的传感器-子载波选择策略，通过优先加权子载波-传感器覆盖（PWSSC）算法实现。在UT-HAR数据集和基于nexmon的真实测试平台上进行评估后，与基线未压缩的CSI数据传输相比，AutoCompress平均减少了4000倍的数据传输，提高了35.5%的网络吞吐量，减少了90%的延迟，同时保持了较高的推理精度。这些结果表明AutoCompress是分布式无线传感系统的可扩展、可解释和带宽效率高的解决方案。

{"title":"AutoCompress: Improving network efficiency for distributed wireless sensing applications","authors":"Yamini Shankar, Ayon Chakraborty","doi":"10.1016/j.pmcj.2026.102178","DOIUrl":"10.1016/j.pmcj.2026.102178","url":null,"abstract":"<div><div>Radio Frequency (RF)-based wireless sensing provides a privacy-preserving and non-intrusive alternative to vision-based systems for human activity recognition and environmental monitoring. However, transmitting high dimensional Channel State Information (CSI) from multiple distributed sensors leads to severe network congestion, especially at scale. We propose <span>AutoCompress</span>, a model-aware framework that intelligently compresses CSI spectrograms by prioritizing the most informative elements. Using Sparse Sensor Placement Optimization for Classification (SSPOC), <span>AutoCompress</span> computes element-wise importance scores for each spectrogram. These scores guide a sensor-subcarrier selection strategy under bandwidth constraints, implemented via the Prioritized Weighted Subcarrier-Sensor Cover (PWSSC) algorithm. Evaluated on the UT-HAR dataset and a real-world Nexmon-based testbed, <span>AutoCompress</span> achieves an average <span><math><mrow><mo>></mo><mn>4000</mn><mo>×</mo></mrow></math></span> reduction in data transmission, improves network throughput by 35.5%, and reduces latency by <span><math><mrow><mo>></mo><mn>90</mn><mtext>%</mtext></mrow></math></span>-all while maintaining high inference accuracy, compared to baseline uncompressed CSI data transmission. These results demonstrate <span>AutoCompress</span> as a scalable, interpretable, and bandwidth-efficient solution for distributed wireless sensing systems.</div></div>","PeriodicalId":49005,"journal":{"name":"Pervasive and Mobile Computing","volume":"117 ","pages":"Article 102178"},"PeriodicalIF":3.5,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146173211","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Design and implementation of a platform for stateful agents at the edge 边缘有状态代理平台的设计和实现

IF 3.5 3区计算机科学 Q2 COMPUTER SCIENCE, INFORMATION SYSTEMS

Pervasive and Mobile Computing

Pub Date : 2026-04-01 Epub Date: 2026-01-28 DOI: 10.1016/j.pmcj.2026.102175

Claudio Cicconetti , Emanuele Carlini , Chen Chen , Roman Kolcun , Richard Mortier

Edge–cloud computing infrastructures are increasingly widespread as they combine the flexibility of cloud-native development tools with the performance and security of distributed computing environments. Function-as-a-Service has emerged as a powerful abstraction that overcomes the limitations of a micro-service architecture. However, it generally does not support stateful functions, making it unsuitable for many practical applications in, e.g., Internet of Things (IoT) and real-time analytics. In this paper, we explore a novel paradigm, based on stateful asynchronous agents, that goes beyond traditional serverless computing. We focus on several key technical aspects: programming model, deployment procedures, design of a flexible compute node, and state management. We illustrate our paradigm using the EDGELESS platform as a concrete implementation of this stateful agents’ pattern. We report proof-of-concept experiment results obtained in a testbed with heterogeneous resource-constrained edge nodes that showcase some distinguishing features of our platform: scalable management of lightweight function instances, the advantage of keeping the state local at function instances, and delegated orchestration to enable a third-party agent to make migration decisions in a group of local nodes.

边缘云计算基础设施越来越广泛，因为它们将云原生开发工具的灵活性与分布式计算环境的性能和安全性相结合。功能即服务已经作为一种强大的抽象出现，它克服了微服务架构的局限性。然而，它通常不支持有状态功能，使其不适合许多实际应用，例如物联网（IoT）和实时分析。在本文中，我们探索了一种基于有状态异步代理的新范式，它超越了传统的无服务器计算。我们关注几个关键的技术方面：编程模型、部署过程、灵活计算节点的设计和状态管理。我们使用无边界平台作为这个有状态代理模式的具体实现来说明我们的范例。我们报告了在具有异构资源约束边缘节点的测试平台中获得的概念验证实验结果，这些结果展示了我们平台的一些显着特性：轻量级功能实例的可扩展管理，在功能实例中保持状态本地的优势，以及委托编排以使第三方代理能够在一组本地节点中做出迁移决策。

{"title":"Design and implementation of a platform for stateful agents at the edge","authors":"Claudio Cicconetti , Emanuele Carlini , Chen Chen , Roman Kolcun , Richard Mortier","doi":"10.1016/j.pmcj.2026.102175","DOIUrl":"10.1016/j.pmcj.2026.102175","url":null,"abstract":"<div><div>Edge–cloud computing infrastructures are increasingly widespread as they combine the flexibility of cloud-native development tools with the performance and security of distributed computing environments. Function-as-a-Service has emerged as a powerful abstraction that overcomes the limitations of a micro-service architecture. However, it generally does not support stateful functions, making it unsuitable for many practical applications in, e.g., Internet of Things (IoT) and real-time analytics. In this paper, we explore a novel paradigm, based on stateful asynchronous agents, that goes beyond traditional serverless computing. We focus on several key technical aspects: programming model, deployment procedures, design of a flexible compute node, and state management. We illustrate our paradigm using the <em>EDGELESS</em> platform as a concrete implementation of this stateful agents’ pattern. We report proof-of-concept experiment results obtained in a testbed with heterogeneous resource-constrained edge nodes that showcase some distinguishing features of our platform: scalable management of lightweight function instances, the advantage of keeping the state local at function instances, and delegated orchestration to enable a third-party agent to make migration decisions in a group of local nodes.</div></div>","PeriodicalId":49005,"journal":{"name":"Pervasive and Mobile Computing","volume":"117 ","pages":"Article 102175"},"PeriodicalIF":3.5,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146079113","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

TCP-HAR: On-Device Transferable and Copyright-Preserving Human Activity Recognition TCP-HAR：设备上可转让和保护版权的人类活动识别

IF 3.5 3区计算机科学 Q2 COMPUTER SCIENCE, INFORMATION SYSTEMS

Pervasive and Mobile Computing

Pub Date : 2026-04-01 Epub Date: 2026-01-13 DOI: 10.1016/j.pmcj.2026.102163

Alessio Sacco , Bruno Palermo , Giulio Figliolino , Chiara Contoli , Guido Marchetto , Flavio Esposito

Teaching a machine to accurately identify human activities from sensor data poses a significant challenge, which is further compounded by considerations of data privacy, resource costs, and responsiveness, particularly within the constraints of devices like smartphones. While current solutions efficiently identify activities, trained models are barely portable in scenarios composed of diverse activities and limited battery life devices, such as smartphones. This paper introduces Transferable and Copyright-Preserving Human Activity Recognition (TCP-HAR), a mobile-based HAR system that integrates digital watermarking, Federated Learning (FL), Transfer Learning (TL), and compression techniques to provide efficient human activity recognition while providing copyright protection of deep neural network models over Android smartphones. Our solution optimizes the utilization of FL, TL, and their combination (FTL) by extensively testing standalone TL models in offline contexts and comparing these results with FL across a network of mobile devices. Our findings highlight the benefits of TCP-HAR for mobile environments in terms of accuracy, F1-score, and training time. In addition, our proposed watermarking mechanism is robust yet computationally efficient, ensuring ownership verification without compromising the scalability of the TFL process.

教机器从传感器数据中准确识别人类活动是一项重大挑战，考虑到数据隐私、资源成本和响应能力，尤其是在智能手机等设备的限制下，这一挑战进一步复杂化。虽然目前的解决方案可以有效地识别活动，但经过训练的模型在由各种活动和有限电池寿命的设备（如智能手机）组成的场景中几乎无法携带。本文介绍了可转移和保护版权的人类活动识别（TCP-HAR），这是一种基于移动的HAR系统，它集成了数字水印、联邦学习（FL）、迁移学习（TL）和压缩技术，在提供有效的人类活动识别的同时，为Android智能手机上的深度神经网络模型提供版权保护。我们的解决方案通过在离线环境中广泛测试独立的TL模型，并将这些结果与移动设备网络中的FL进行比较，优化了FL、TL及其组合（FTL）的利用率。我们的研究结果强调了TCP-HAR在准确性、f1分数和训练时间方面对移动环境的好处。此外，我们提出的水印机制鲁棒且计算效率高，确保所有权验证而不影响TFL过程的可扩展性。

{"title":"TCP-HAR: On-Device Transferable and Copyright-Preserving Human Activity Recognition","authors":"Alessio Sacco , Bruno Palermo , Giulio Figliolino , Chiara Contoli , Guido Marchetto , Flavio Esposito","doi":"10.1016/j.pmcj.2026.102163","DOIUrl":"10.1016/j.pmcj.2026.102163","url":null,"abstract":"<div><div>Teaching a machine to accurately identify human activities from sensor data poses a significant challenge, which is further compounded by considerations of data privacy, resource costs, and responsiveness, particularly within the constraints of devices like smartphones. While current solutions efficiently identify activities, trained models are barely portable in scenarios composed of diverse activities and limited battery life devices, such as smartphones. This paper introduces Transferable and Copyright-Preserving Human Activity Recognition (TCP-HAR), a mobile-based HAR system that integrates digital watermarking, Federated Learning (FL), Transfer Learning (TL), and compression techniques to provide efficient human activity recognition while providing copyright protection of deep neural network models over Android smartphones. Our solution optimizes the utilization of FL, TL, and their combination (FTL) by extensively testing standalone TL models in offline contexts and comparing these results with FL across a network of mobile devices. Our findings highlight the benefits of TCP-HAR for mobile environments in terms of accuracy, F1-score, and training time. In addition, our proposed watermarking mechanism is robust yet computationally efficient, ensuring ownership verification without compromising the scalability of the TFL process.</div></div>","PeriodicalId":49005,"journal":{"name":"Pervasive and Mobile Computing","volume":"117 ","pages":"Article 102163"},"PeriodicalIF":3.5,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146079112","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Traffic analysis and resource adaptation in large-scale 5G multi-layer edge networks 大规模5G多层边缘网络的流量分析与资源适配

IF 3.5 3区计算机科学 Q2 COMPUTER SCIENCE, INFORMATION SYSTEMS

Pervasive and Mobile Computing

Pub Date : 2026-03-01 Epub Date: 2025-12-27 DOI: 10.1016/j.pmcj.2025.102158

Marcello Pietri , Natalia Selini Hadjidimitriou , Marco Mamei , Marco Picone , Enrico Rossini , Edoardo Maria Sanna , Jovanka Adzic , Andrea Buldorini

In this research, we propose automating network management through data-driven intelligence, with a particular focus on anomalies and network traffic during specific events or periods. We analyze a large dataset collected by Orange mobile network operator in France with the goal of forecasting mobile demand for different classes of services. To model the underlying network infrastructure, we introduce a model for the underlying network based on a hierarchy of virtualization layers and slices. Building on this model, we propose algorithms to optimize the resources allocated to network slices and traffic distribution within the operator’s network. Network performance is evaluated as the fraction of time the mobile traffic is within the capacity of the network. Our results demonstrate that dynamic reallocation of resources among slices, and dynamic load balancing (traffic shaping) between nodes notably improves network performance. These results provide insights into critical aspects related to future 5G network management.

在这项研究中，我们建议通过数据驱动的智能自动化网络管理，特别关注特定事件或时期的异常和网络流量。我们分析了法国Orange移动网络运营商收集的大型数据集，目的是预测不同类别服务的移动需求。为了对底层网络基础设施进行建模，我们引入了一个基于虚拟化层和切片层次结构的底层网络模型。在此模型的基础上，我们提出了优化分配给网络切片的资源和运营商网络内流量分配的算法。网络性能是用移动流量在网络容量范围内的时间比例来评估的。我们的结果表明，在片之间动态重新分配资源，以及节点之间的动态负载平衡（流量整形）显着提高了网络性能。这些结果为未来5G网络管理的关键方面提供了见解。

{"title":"Traffic analysis and resource adaptation in large-scale 5G multi-layer edge networks","authors":"Marcello Pietri , Natalia Selini Hadjidimitriou , Marco Mamei , Marco Picone , Enrico Rossini , Edoardo Maria Sanna , Jovanka Adzic , Andrea Buldorini","doi":"10.1016/j.pmcj.2025.102158","DOIUrl":"10.1016/j.pmcj.2025.102158","url":null,"abstract":"<div><div>In this research, we propose automating network management through data-driven intelligence, with a particular focus on anomalies and network traffic during specific events or periods. We analyze a large dataset collected by Orange mobile network operator in France with the goal of forecasting mobile demand for different classes of services. To model the underlying network infrastructure, we introduce a model for the underlying network based on a hierarchy of virtualization layers and slices. Building on this model, we propose algorithms to optimize the resources allocated to network slices and traffic distribution within the operator’s network. Network performance is evaluated as the fraction of time the mobile traffic is within the capacity of the network. Our results demonstrate that dynamic reallocation of resources among slices, and dynamic load balancing (traffic shaping) between nodes notably improves network performance. These results provide insights into critical aspects related to future 5G network management.</div></div>","PeriodicalId":49005,"journal":{"name":"Pervasive and Mobile Computing","volume":"116 ","pages":"Article 102158"},"PeriodicalIF":3.5,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145884604","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Sybil-aware adaptive defence framework for robust federated learning 用于健壮联邦学习的sybil感知自适应防御框架

IF 3.5 3区计算机科学 Q2 COMPUTER SCIENCE, INFORMATION SYSTEMS

Pervasive and Mobile Computing

Pub Date : 2026-03-01 Epub Date: 2025-12-16 DOI: 10.1016/j.pmcj.2025.102157

Dnyanesh Khedekar , Tanmaya Mahapatra , Amitesh Singh Rajput

Federated Learning (FL) enables distributed learning of sensitive and multi-location data to enhance data privacy and avoid data leakages. In FL, participating clients only share the trained local model updates to the external server for global aggregation. However, FL is vulnerable to security threats like adaptive data poisoning attacks in non-IID data scenarios, especially coordinated attacks by colluding sybils. Existing defences struggle against sybils as their focus has been on either analysing model behaviour or strengthening aggregation mechanisms. Sybils can collude by sharing their model updates with each other in the external environment. They can then strategically manipulate and share poisoned model updates for global aggregation, specifically employing label-flipping attacks. This paper introduces a novel defence framework that shifts the focus from model analysis to sybil behaviour analysis utilizing historical pairwise cosine similarity of client updates during the training process. By establishing a dynamic threshold and analysing patterns of similarity change among participating clients, the proposed defence framework detects sybils exhibiting coordinated data poisoning and excludes them from the subsequent global aggregation process. This approach is adaptable to various underlying aggregation methods, providing a robust defence against collusive data poisoning attacks and improving model resilience and convergence even under challenging non-IID settings. Comprehensive evaluations demonstrate a significant reduction in the Attack Success Ratio, i.e. from over 40% to below 1%, showcasing its superior efficacy compared to state-of-the-art defences against targeted data poisoning attacks.

联邦学习（Federated Learning， FL）可以对敏感数据和多位置数据进行分布式学习，以增强数据的隐私性，避免数据泄露。在FL中，参与的客户端仅将经过训练的本地模型更新共享到外部服务器以进行全局聚合。然而，在非iid数据场景下，FL容易受到自适应数据中毒攻击等安全威胁，特别是由串通黑客发起的协同攻击。现有的防御与sysyls斗争，因为它们的重点要么是分析模型行为，要么是加强聚合机制。sybil可以通过在外部环境中相互共享他们的模型更新来串通起来。然后，他们可以策略性地操纵和共享有毒的模型更新，用于全局聚合，特别是使用标签翻转攻击。本文介绍了一种新的防御框架，该框架利用客户端在训练过程中更新的历史两两余弦相似性，将重点从模型分析转移到符号行为分析。通过建立动态阈值和分析参与客户端之间的相似性变化模式，所提出的防御框架检测出表现出协调数据中毒的sybils，并将其排除在随后的全局聚合过程之外。这种方法适用于各种底层聚合方法，提供了对串通数据中毒攻击的强大防御，并提高了模型的弹性和收敛性，即使在具有挑战性的非iid设置下也是如此。综合评估表明，攻击成功率显著降低，即从40%以上降至1%以下，与最先进的针对目标数据中毒攻击的防御相比，显示出其优越的功效。

{"title":"Sybil-aware adaptive defence framework for robust federated learning","authors":"Dnyanesh Khedekar , Tanmaya Mahapatra , Amitesh Singh Rajput","doi":"10.1016/j.pmcj.2025.102157","DOIUrl":"10.1016/j.pmcj.2025.102157","url":null,"abstract":"<div><div>Federated Learning (FL) enables distributed learning of sensitive and multi-location data to enhance data privacy and avoid data leakages. In FL, participating clients only share the trained local model updates to the external server for global aggregation. However, FL is vulnerable to security threats like adaptive data poisoning attacks in non-IID data scenarios, especially coordinated attacks by colluding sybils. Existing defences struggle against sybils as their focus has been on either analysing model behaviour or strengthening aggregation mechanisms. Sybils can collude by sharing their model updates with each other in the external environment. They can then strategically manipulate and share poisoned model updates for global aggregation, specifically employing label-flipping attacks. This paper introduces a novel defence framework that shifts the focus from model analysis to sybil behaviour analysis utilizing historical pairwise cosine similarity of client updates during the training process. By establishing a dynamic threshold and analysing patterns of similarity change among participating clients, the proposed defence framework detects sybils exhibiting coordinated data poisoning and excludes them from the subsequent global aggregation process. This approach is adaptable to various underlying aggregation methods, providing a robust defence against collusive data poisoning attacks and improving model resilience and convergence even under challenging non-IID settings. Comprehensive evaluations demonstrate a significant reduction in the Attack Success Ratio, i.e. from over 40% to below 1%, showcasing its superior efficacy compared to state-of-the-art defences against targeted data poisoning attacks.</div></div>","PeriodicalId":49005,"journal":{"name":"Pervasive and Mobile Computing","volume":"116 ","pages":"Article 102157"},"PeriodicalIF":3.5,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145790984","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A self-adaptive framework for child healthcare in IoT environment using AI-based prediction 基于人工智能预测的物联网环境下儿童医疗自适应框架

IF 3.5 3区计算机科学 Q2 COMPUTER SCIENCE, INFORMATION SYSTEMS

Pervasive and Mobile Computing

Pub Date : 2026-03-01 Epub Date: 2025-12-04 DOI: 10.1016/j.pmcj.2025.102153

Euijong Lee , Jaemin Jeong , Gyuchan Jo , Taegyeom Lee , Gee-Myung Moon , Young-Duk Seo , Ji-Hoon Jeong

Childhood overweight and obesity have emerged as some of the most serious global public health challenges, as they can lead to a variety of health-related problems and the early development of chronic diseases. The healthcare domain has been transformed by the integration of the Internet of Things (IoT), leading to the development of digital healthcare solutions. This integration has led to an increase in the health data collected from a variety of IoT sources. Consequently, advanced technologies are required to analyze health data, and Artificial Intelligence (AI) has been employed to extract meaningful insights from the data. Moreover, these technologies can be effectively applied in healthcare to address childhood overweight and obesity. A self-adaptive framework is proposed to manage childhood weight using lifelog data from IoT environments. An ensemble-based learning model is applied to predict weight using the lifelog data. A smartphone application providing real-world services was developed, and lifelog data were collected from a cohort of 362 children aged 104 to 152 months. Reasonable results were obtained, indicating that the proposed ensemble model can predict childhood weight change using lifelog data with an accuracy of 0.9711 and an F1-score 0.9725. Also, the results indicated that appropriate rewards can encourage human involvement and positively influence data quality.The results demonstrated the efficiency of the proposed framework with human involvement in weight prediction. The experimental results demonstrate the efficiency of the proposed framework and its potential application in healthcare services to enhance children’s health.

儿童超重和肥胖已成为一些最严重的全球公共卫生挑战，因为它们可导致各种与健康有关的问题和慢性病的早期发展。物联网（IoT）的集成已经改变了医疗保健领域，从而导致了数字医疗保健解决方案的发展。这种整合导致了从各种物联网来源收集的健康数据的增加。因此，需要先进的技术来分析健康数据，并使用人工智能（AI）从数据中提取有意义的见解。此外，这些技术可以有效地应用于医疗保健，以解决儿童超重和肥胖问题。提出了一种自适应框架，利用物联网环境中的生活日志数据来管理儿童体重。采用基于集成的学习模型，利用生活日志数据预测体重。研究人员开发了一款提供真实世界服务的智能手机应用程序，并收集了362名年龄在104至152个月之间的儿童的生活日志数据。结果表明，本文提出的集成模型可以利用生活日志数据预测儿童体重变化，准确率为0.9711，f1得分为0.9725。此外，结果表明，适当的奖励可以鼓励人的参与，并积极影响数据质量。结果表明，在人类参与权重预测的情况下，所提出的框架是有效的。实验结果证明了该框架的有效性及其在医疗保健服务中提高儿童健康水平的潜在应用。

{"title":"A self-adaptive framework for child healthcare in IoT environment using AI-based prediction","authors":"Euijong Lee , Jaemin Jeong , Gyuchan Jo , Taegyeom Lee , Gee-Myung Moon , Young-Duk Seo , Ji-Hoon Jeong","doi":"10.1016/j.pmcj.2025.102153","DOIUrl":"10.1016/j.pmcj.2025.102153","url":null,"abstract":"<div><div>Childhood overweight and obesity have emerged as some of the most serious global public health challenges, as they can lead to a variety of health-related problems and the early development of chronic diseases. The healthcare domain has been transformed by the integration of the Internet of Things (IoT), leading to the development of digital healthcare solutions. This integration has led to an increase in the health data collected from a variety of IoT sources. Consequently, advanced technologies are required to analyze health data, and Artificial Intelligence (AI) has been employed to extract meaningful insights from the data. Moreover, these technologies can be effectively applied in healthcare to address childhood overweight and obesity. A self-adaptive framework is proposed to manage childhood weight using lifelog data from IoT environments. An ensemble-based learning model is applied to predict weight using the lifelog data. A smartphone application providing real-world services was developed, and lifelog data were collected from a cohort of 362 children aged 104 to 152 months. Reasonable results were obtained, indicating that the proposed ensemble model can predict childhood weight change using lifelog data with an accuracy of 0.9711 and an F1-score 0.9725. Also, the results indicated that appropriate rewards can encourage human involvement and positively influence data quality.The results demonstrated the efficiency of the proposed framework with human involvement in weight prediction. The experimental results demonstrate the efficiency of the proposed framework and its potential application in healthcare services to enhance children’s health.</div></div>","PeriodicalId":49005,"journal":{"name":"Pervasive and Mobile Computing","volume":"116 ","pages":"Article 102153"},"PeriodicalIF":3.5,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145738157","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

The structure design of the smart sock prototype integrated with stretchable hybrid electronic temperature sensing yarn for real-time temperature monitoring 结合可拉伸混合电子感温纱进行实时温度监测的智能袜子原型的结构设计

IF 3.5 3区计算机科学 Q2 COMPUTER SCIENCE, INFORMATION SYSTEMS

Pervasive and Mobile Computing

Pub Date : 2026-03-01 Epub Date: 2025-11-26 DOI: 10.1016/j.pmcj.2025.102136

Sumonta Ghosh , Fenye Meng , Rony Shaha , Jiyong Hu

Rapid advances in electronic textiles have enabled the development of smart socks with temperature-sensing capabilities for real-time foot temperature monitoring, where stretchability, user comfort, durability, bending, washability, sensing location error during wear, and higher manufacturing cost are drawbacks. This study introduces a smart sock prototype that integrates highly stretchable hybrid electronic temperature-sensing (SHETS) yarns in key physiological foot regions—hallux, metatarsal, midfoot, and heel—to detect temperature variations and address the challenges of existing and commercial products. Structural design of SHETS yarns involved wrapping, novel interconnection, encapsulation, and braiding techniques to integrate miniature thermistor within the yarn structure, showing high stretchability, durability, washability, and sensing accuracy. Low-power microcontroller transmits analog data to a digital format, and the web-based interface enables users to monitor data through mobile phone applications in real-time. The prototype demonstrates high accuracy, durability, and reliability under various conditions, with average temperature error ranging from ±0.23 °C to ±0.27 °C. The prototype maintains durability and stability in hot, cold, and sweaty conditions. Physical activities like walking, running, and cycling demonstrate the durability and stability of foot temperature changes, while extended wear shows low power consumption and stability. The device withstands over 30 washing cycles with minimal accuracy loss (maximum ±0.28 °C error) and retains the mechanical and electrical properties of SHETS yarn under repeated stretching and bending. Additionally, the android-based intelligent foot alert system enhances usability by providing real-time monitoring and alerts on smartphones, offering a cost-effective, energy-efficient, and user-friendly solution for proactive foot health management. Available code: https://github.com/Sumonta-e-textile/Smart-Materials-and-Electronic-Textiles-Lab.

电子纺织品的快速发展使得具有温度传感功能的智能袜子能够进行实时足部温度监测，其中拉伸性，用户舒适性，耐用性，弯曲性，可洗涤性，磨损时的传感位置误差以及较高的制造成本是缺点。本研究介绍了一种智能袜子原型，该原型将高度可拉伸的混合电子温度传感（SHETS）纱线集成在脚的关键生理区域-拇趾，跖骨，足中部和脚跟-以检测温度变化并解决现有和商业产品的挑战。shts纱线的结构设计采用包绕、新型互连、封装和编织技术，将微型热敏电阻集成到纱线结构中，具有高拉伸性、耐久性、耐洗性和传感精度。低功耗微控制器将模拟数据传输为数字格式，基于web的界面使用户可以通过手机应用程序实时监控数据。该样机在各种条件下具有高精度、耐用性和可靠性，平均温度误差范围为±0.23°C至±0.27°C。原型在热、冷和出汗的条件下保持耐用性和稳定性。步行、跑步和骑自行车等体育活动表现出足部温度变化的耐久性和稳定性，而长时间穿着则表现出低功耗和稳定性。该设备可承受超过30次洗涤循环，精度损失最小（最大±0.28°C误差），并在重复拉伸和弯曲下保持SHETS纱线的机械和电气性能。此外，基于android的智能足部警报系统通过在智能手机上提供实时监控和警报，提高了可用性，为主动足部健康管理提供了经济高效、节能和用户友好的解决方案。可用代码：https://github.com/Sumonta-e-textile/Smart-Materials-and-Electronic-Textiles-Lab。

{"title":"The structure design of the smart sock prototype integrated with stretchable hybrid electronic temperature sensing yarn for real-time temperature monitoring","authors":"Sumonta Ghosh , Fenye Meng , Rony Shaha , Jiyong Hu","doi":"10.1016/j.pmcj.2025.102136","DOIUrl":"10.1016/j.pmcj.2025.102136","url":null,"abstract":"<div><div>Rapid advances in electronic textiles have enabled the development of smart socks with temperature-sensing capabilities for real-time foot temperature monitoring, where stretchability, user comfort, durability, bending, washability, sensing location error during wear, and higher manufacturing cost are drawbacks. This study introduces a smart sock prototype that integrates highly stretchable hybrid electronic temperature-sensing (SHETS) yarns in key physiological foot regions—hallux, metatarsal, midfoot, and heel—to detect temperature variations and address the challenges of existing and commercial products. Structural design of SHETS yarns involved wrapping, novel interconnection, encapsulation, and braiding techniques to integrate miniature thermistor within the yarn structure, showing high stretchability, durability, washability, and sensing accuracy. Low-power microcontroller transmits analog data to a digital format, and the web-based interface enables users to monitor data through mobile phone applications in real-time. The prototype demonstrates high accuracy, durability, and reliability under various conditions, with average temperature error ranging from ±0.23 °C to ±0.27 °C. The prototype maintains durability and stability in hot, cold, and sweaty conditions. Physical activities like walking, running, and cycling demonstrate the durability and stability of foot temperature changes, while extended wear shows low power consumption and stability. The device withstands over 30 washing cycles with minimal accuracy loss (maximum ±0.28 °C error) and retains the mechanical and electrical properties of SHETS yarn under repeated stretching and bending. Additionally, the android-based intelligent foot alert system enhances usability by providing real-time monitoring and alerts on smartphones, offering a cost-effective, energy-efficient, and user-friendly solution for proactive foot health management. Available code: <span><span>https://github.com/Sumonta-e-textile/Smart-Materials-and-Electronic-Textiles-Lab</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":49005,"journal":{"name":"Pervasive and Mobile Computing","volume":"116 ","pages":"Article 102136"},"PeriodicalIF":3.5,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145659149","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

MDWD-KAN: Multilevel discrete wavelet decomposition with Kolmogorov–Arnold network for fall detection and activity recognition using wearable sensors 基于Kolmogorov-Arnold网络的多电平离散小波分解可穿戴传感器跌倒检测和活动识别

IF 3.5 3区计算机科学 Q2 COMPUTER SCIENCE, INFORMATION SYSTEMS

Pervasive and Mobile Computing

Pub Date : 2026-03-01 Epub Date: 2026-01-06 DOI: 10.1016/j.pmcj.2026.102160

Zhiyuan Jiang , Sike Ni , Mohammed A.A. Al-qaness

Fall detection and Human Activity Recognition (HAR) are crucial applications in pervasive and mobile computing, enabling real-time monitoring of individuals – especially the elderly or patients – for enhanced safety and health management. Wearable devices have emerged as a critical tool for continuous activity monitoring, enabling real-time detection and intervention. However, the quality of data collected by wearable sensors faces several challenges, including noise interference, instability due to wearing, and individual differences. To address these challenges, this paper proposes a feature stepwise fusion detection system based on a multilevel discrete wavelet decomposition with Kolmogorov–Arnold Network, namely MDWD-KAN. This model utilizes multilevel wavelet decomposition to perform multiresolution analysis on sensor signals, extracting multilevel features and effectively enhancing feature stability and noise resistance. Additionally, through a heterogeneous model and a multilevel feature fusion strategy, MDWD-KAN achieves complementary low-frequency and high-frequency features, improving the recognition capability for complex motion patterns. Experiments were conducted on three public datasets: MobiAct, SisFall, and UniMiB-SHAR. The results show that MDWD-KAN achieves average recognition accuracies of 99.67%, 99.90%, and 99.65%, respectively, for binary classification (fall and non-fall), and 98.85%, 85.45%, and 96.86%, respectively, for multiclassification.

跌倒检测和人体活动识别（HAR）是普及和移动计算中的关键应用，能够实时监测个人，特别是老年人或患者，以加强安全和健康管理。可穿戴设备已经成为持续活动监测的关键工具，可以实现实时检测和干预。然而，可穿戴传感器收集的数据质量面临着一些挑战，包括噪声干扰、佩戴不稳定以及个体差异。为了解决这些问题，本文提出了一种基于Kolmogorov-Arnold网络的多层离散小波分解的特征逐步融合检测系统MDWD-KAN。该模型利用多级小波分解对传感器信号进行多分辨率分析，提取多级特征，有效增强特征稳定性和抗噪性。此外，MDWD-KAN通过异构模型和多层次特征融合策略，实现低频和高频特征的互补，提高了对复杂运动模式的识别能力。实验在三个公共数据集上进行：MobiAct、SisFall和unimib - share。结果表明，mddd - kan对二分类（跌倒和非跌倒）的平均识别准确率分别为99.67%、99.90%和99.65%，对多分类的平均识别准确率分别为98.85%、85.45%和96.86%。

{"title":"MDWD-KAN: Multilevel discrete wavelet decomposition with Kolmogorov–Arnold network for fall detection and activity recognition using wearable sensors","authors":"Zhiyuan Jiang , Sike Ni , Mohammed A.A. Al-qaness","doi":"10.1016/j.pmcj.2026.102160","DOIUrl":"10.1016/j.pmcj.2026.102160","url":null,"abstract":"<div><div>Fall detection and Human Activity Recognition (HAR) are crucial applications in pervasive and mobile computing, enabling real-time monitoring of individuals – especially the elderly or patients – for enhanced safety and health management. Wearable devices have emerged as a critical tool for continuous activity monitoring, enabling real-time detection and intervention. However, the quality of data collected by wearable sensors faces several challenges, including noise interference, instability due to wearing, and individual differences. To address these challenges, this paper proposes a feature stepwise fusion detection system based on a multilevel discrete wavelet decomposition with Kolmogorov–Arnold Network, namely MDWD-KAN. This model utilizes multilevel wavelet decomposition to perform multiresolution analysis on sensor signals, extracting multilevel features and effectively enhancing feature stability and noise resistance. Additionally, through a heterogeneous model and a multilevel feature fusion strategy, MDWD-KAN achieves complementary low-frequency and high-frequency features, improving the recognition capability for complex motion patterns. Experiments were conducted on three public datasets: MobiAct, SisFall, and UniMiB-SHAR. The results show that MDWD-KAN achieves average recognition accuracies of 99.67%, 99.90%, and 99.65%, respectively, for binary classification (fall and non-fall), and 98.85%, 85.45%, and 96.86%, respectively, for multiclassification.</div></div>","PeriodicalId":49005,"journal":{"name":"Pervasive and Mobile Computing","volume":"116 ","pages":"Article 102160"},"PeriodicalIF":3.5,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145925535","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0