Internet of Things最新文献_第2页

Towards dynamic cooperative scheduling in multi-UAV urban sensing: A hierarchical reinforcement learning approach 多无人机城市感知中的动态协同调度：层次强化学习方法

IF 7.1 3区计算机科学

Internet of Things Pub Date : 2026-07-01 Epub Date: 2026-06-05 DOI: 10.1016/j.iot.2026.101985

Mengge Jiao, Kaimin Wei, Zhiquan Liu, Jinpeng Chen, Chengkun Jia, Yan Li, Tingrui Pei

{"title":"Towards dynamic cooperative scheduling in multi-UAV urban sensing: A hierarchical reinforcement learning approach","authors":"Mengge Jiao, Kaimin Wei, Zhiquan Liu, Jinpeng Chen, Chengkun Jia, Yan Li, Tingrui Pei","doi":"10.1016/j.iot.2026.101985","DOIUrl":"10.1016/j.iot.2026.101985","url":null,"abstract":"<div><div>Multi-UAV systems are increasingly deployed in urban sensing such as traffic surveillance and environmental monitoring. However, as task scales expand and environmental dynamics intensify, existing schedulers struggle to balance global efficiency with local real-time performance when confronted with semantically ambiguous tasks, temporal conflicts, and rapidly growing decision spaces. To address this, we propose HRL-DCSF, a hierarchical reinforcement learning-driven dynamic cooperative scheduling framework that achieves end-to-end optimisation from natural-language-style directives to multi-UAV execution. HRL-DCSF introduces a unified three-layer architecture comprising task decomposition, task allocation, and task execution. We formulate the entire pipeline as a hierarchical Markov decision process (HiMDP): the allocation layer is modeled as a semi-Markov decision process (SMDP) for temporally extended planning, while the execution layer is modeled as a constrained Markov decision process (CMDP) to explicitly account for energy, communication, and storage constraints. To enhance adaptivity in dynamic environments, HRL-DCSF further integrates a dynamic cluster-formation mechanism, periodic load rebalancing, and an alternating hierarchical training (AHT) algorithm that synchronises cross-layer policies for stable learning. Experimental results demonstrate that in a small-scale convergence test with four UAVs and eighty tasks against DL-DRL, the converged total task value is higher by approximately 46% relative to DL-DRL. In the large-scale six-UAV multi-baseline study at <span><math><mrow><mi>N</mi><mo>=</mo><mn>1000</mn></mrow></math></span>, HRL-DCSF attains a mean resource utilisation of 84.0%.</div></div>","PeriodicalId":29968,"journal":{"name":"Internet of Things","volume":"38 ","pages":"Article 101985"},"PeriodicalIF":7.1,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148523718","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

MEGA-CoPLA: Mixture of experts enhanced generative adversarial network-based collaborative physical layer authentication for internet of vehicles MEGA-CoPLA：混合专家增强的基于生成对抗网络的车辆互联网协同物理层身份验证

IF 7.1 3区计算机科学

Internet of Things Pub Date : 2026-07-01 Epub Date: 2026-06-24 DOI: 10.1016/j.iot.2026.102009

Mubarak Umar, Huang Xin, Yuanying Qu, Mahmud Yusuf Ahmed, Hafsa Kabir Ahmad

{"title":"MEGA-CoPLA: Mixture of experts enhanced generative adversarial network-based collaborative physical layer authentication for internet of vehicles","authors":"Mubarak Umar, Huang Xin, Yuanying Qu, Mahmud Yusuf Ahmed, Hafsa Kabir Ahmad","doi":"10.1016/j.iot.2026.102009","DOIUrl":"10.1016/j.iot.2026.102009","url":null,"abstract":"<div><div>Conventional physical layer authentication (PLA) techniques that rely on a single device suffer from poor performance due to single-point failures and scarce training data, especially in rapidly changing Internet of Vehicles (IoV) communication environments. Although collaborative PLA (CPLA) methods mitigate these challenges through distributed authentication and Generative Adversarial Networks (GANs)-based data augmentation, they suffer from limited adaptability, resulting in weak generalization and degraded performance in unseen environments. Additionally, the absence of personalized learning prevents authentication models from adapting to each vehicle’s unique channel conditions. To address these limitations, we propose MEGA-CoPLA, a Mixture of Experts (MoE) enhanced GAN-based CPLA framework with Gossip Learning (GL). Each vehicle employs a local GAN to generate synthetic fingerprints and enrich its training data for improved authentication. To enhance adaptability across diverse propagation conditions, we introduce a MoE architecture into the authentication framework to provide context-aware authentication by activating only the most relevant experts for each scenario, thereby improving generalization and authentication performance under dynamic IoV environments. Gossip learning, which enables peer-to-peer model sharing and personalized learning, is then used to collaboratively train Convolutional Neural Network (CNN) models and authenticate transmitters through Channel State Information (CSI) and 2-D location data. Simulation results using the dataset generated from the Quasi Deterministic Radio Channel Generator (QuaDRiGa) framework demonstrate that our approach strengthens authentication under dynamic IoV conditions.</div></div>","PeriodicalId":29968,"journal":{"name":"Internet of Things","volume":"38 ","pages":"Article 102009"},"PeriodicalIF":7.1,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148499264","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

Advancements in scalable smart city implementations: The JRC digital platform and its vertical applications 可扩展智慧城市实施的进展：JRC数字平台及其垂直应用

IF 7.1 3区计算机科学

Internet of Things Pub Date : 2026-07-01 Epub Date: 2026-06-19 DOI: 10.1016/j.iot.2026.102004

Flavio Bono, Giovanni Vaglica, Graziano Renaldi, Dario Cazzato, Elena Villares Ojea, José Ricardo Passos Moreira De Sousa

{"title":"Advancements in scalable smart city implementations: The JRC digital platform and its vertical applications","authors":"Flavio Bono, Giovanni Vaglica, Graziano Renaldi, Dario Cazzato, Elena Villares Ojea, José Ricardo Passos Moreira De Sousa","doi":"10.1016/j.iot.2026.102004","DOIUrl":"10.1016/j.iot.2026.102004","url":null,"abstract":"<div><div>Smart city implementations are increasing in the number of services and computational demands requiring solutions that are not only scalable and flexible but also enable the integration of a diverse array of devices, systems, and data sources. This paper expands upon previous research by presenting the smart city platform implemented at the European Commission’s Joint Research Centre in Italy, which is designed to integrate diverse vertical applications such as smart buildings and infrastructures, with several heterogeneous devices and diverse communication systems. Load balancing tests were performed across various configurations to identify the optimal architecture that ensures platform scalability and performance under varying loads. The IT architecture of the platform is therefore designed for future growth, facilitating the future seamless incorporation of new sensors and vertical applications. An important feature of the platform is the integration of digital twins, here applied to individual infrastructures but scalable to entire cities, to enable real-time monitoring, simulations, and predictive analytics. The article discusses the design and implementation of the platform, the scalability and flexibility of the IT architecture, and presents the most representative case studies that demonstrate the platform’s effectiveness in integrating multiple monitoring systems, vertical applications, and digital twins. Furthermore, the proposed platform can efficiently manage large volumes of data from heterogeneous sources, providing real-time analytics, and supports data-driven solutions that can be replicated in real scenarios to support urban management decision-making.</div></div>","PeriodicalId":29968,"journal":{"name":"Internet of Things","volume":"38 ","pages":"Article 102004"},"PeriodicalIF":7.1,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148499265","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A bidirectional cooperative adaptive quantization-based physical layer key generation scheme for vehicular internet of things 基于双向协作自适应量化的车载物联网物理层密钥生成方案

IF 7.1 3区计算机科学

Internet of Things Pub Date : 2026-07-01 Epub Date: 2026-06-14 DOI: 10.1016/j.iot.2026.102002

Jianpo Li, Danni Li, Qi Xu, Yuehua Yang

{"title":"A bidirectional cooperative adaptive quantization-based physical layer key generation scheme for vehicular internet of things","authors":"Jianpo Li, Danni Li, Qi Xu, Yuehua Yang","doi":"10.1016/j.iot.2026.102002","DOIUrl":"10.1016/j.iot.2026.102002","url":null,"abstract":"<div><div>In an intelligent transportation system based on the Internet of Things, ensuring secure communication in Vehicular Ad Hoc Networks (VANETs) is a critical imperative. This paper proposes a Physical Layer Key Generation (PLKG) scheme for VANETs, which achieves highly reliable key generation with low resource overhead. Its main innovations are reflected in two aspects. The first is the Decorrelation-Perturbation Fusion Random Source (DPF-RS) method, which introduces the K-L transform and adaptive fusion into channel-feature processing to enhance randomness and security effectively. The second is the Adaptive Channel Quantization with Guard Band and Bidirectional Cooperation (ACQG-BC) algorithm, which uses kernel density estimation to dynamically optimize quantization parameters, prioritizing reducing the Key Disagreement Rate (KDR) with minimal bidirectional feedback. Simulation results show that the DPF-RS method outperforms baseline methods relying solely on amplitude or phase in terms of Normalized Entropy and Autocorrelation Elimination Rate (AER). Compared with the LDPC-SD, Interpolate-CQG, and CIR-PV algorithms, the ACQG-BC algorithm reduces the KDR by at least 16%, 33%, and 66.7%, respectively. Meanwhile, compared with the LDPC-SD and Interpolate-CQG algorithms, it reduces computational complexity by at least 7.6% and memory usage by at least 11.4%. Furthermore, the proposed PLKG scheme demonstrates robust adaptability in three typical scenarios. It can effectively resist four types of attacks. It has also passed all applicable tests in NIST SP 800-22.</div></div>","PeriodicalId":29968,"journal":{"name":"Internet of Things","volume":"38 ","pages":"Article 102002"},"PeriodicalIF":7.1,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148499376","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

AoI-GTW-aware multi-UAV data collection for time-sensitive internet of things systems 面向时间敏感物联网系统的aoi - gtw感知多无人机数据采集

IF 7.1 3区计算机科学

Internet of Things Pub Date : 2026-07-01 Epub Date: 2026-06-22 DOI: 10.1016/j.iot.2026.102006

Jiaming Lu, Xiaojian Yuan, Chih-Min Yu, Rui Wang

{"title":"AoI-GTW-aware multi-UAV data collection for time-sensitive internet of things systems","authors":"Jiaming Lu, Xiaojian Yuan, Chih-Min Yu, Rui Wang","doi":"10.1016/j.iot.2026.102006","DOIUrl":"10.1016/j.iot.2026.102006","url":null,"abstract":"<div><div>Unmanned aerial vehicles (UAVs) are increasingly deployed as mobile data collectors in Internet of Things (IoT) systems where the value of sensed information depends on both freshness and service-time preference. Existing UAV-IoT planning studies usually represent timeliness through age of information (AoI), deadlines, or mission completion time, while soft preferred service intervals are less explicitly modeled. This paper studies a shared-depot, multi-UAV, multi-sortie data-collection problem that jointly considers AoI and grey time window (GTW) preferences. A tri-objective mixed-integer model is formulated to minimize operational cost, AoI at sampling, and GTW penalty under once-per-node service, sortie endurance, charging linkage, and bounded waiting constraints. To solve the resulting problem, a problem-tailored evolutionary algorithm, MSD-AG-NSGA2, is developed with compact encoding, sortie-aware decoding, dual-biased initialization, targeted structural mutation, and two-layer local search. Experiments on synthetic IoT deployments and adapted Solomon layouts compare the method with four evolutionary baselines and a multi-objective ALNS competitor. Across the main test suite, MSD-AG-NSGA2 improves hypervolume by 39.5% over the strongest generic baseline and by 53.4% over the routing-specialized competitor, while reducing IGD+ by 71.7% and 72.5%, respectively. Operational, ablation, and sensitivity analyses further show that the gains correspond to feasible schedules and interpretable trade-offs among cost, freshness, and GTW satisfaction. The results indicate that combining AoI and GTW preferences provides a useful modeling basis for time-sensitive UAV-assisted IoT data collection.</div></div>","PeriodicalId":29968,"journal":{"name":"Internet of Things","volume":"38 ","pages":"Article 102006"},"PeriodicalIF":7.1,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148499381","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

CornSentry: A multimodal data fusion framework designed for edge-aware corn disease detection 玉米哨兵：设计用于边缘感知玉米病害检测的多模态数据融合框架

IF 7.1 3区计算机科学

Internet of Things Pub Date : 2026-07-01 Epub Date: 2026-06-25 DOI: 10.1016/j.iot.2026.102007

Antonio Guerrero-Ibañez, Juan Contreras-Castillo, Ismael Amezcua-Valdovinos, Sherali Zeadally, Ebtesam Abdullah Al-Suhaimi

{"title":"CornSentry: A multimodal data fusion framework designed for edge-aware corn disease detection","authors":"Antonio Guerrero-Ibañez, Juan Contreras-Castillo, Ismael Amezcua-Valdovinos, Sherali Zeadally, Ebtesam Abdullah Al-Suhaimi","doi":"10.1016/j.iot.2026.102007","DOIUrl":"10.1016/j.iot.2026.102007","url":null,"abstract":"<div><div>Timely and accurate detection of foliar diseases is critical to protect corn production against severe yield losses. To address the limitations of traditional image-only diagnostics, we propose CornSentry, a novel edge-aware framework driven by multimodal data fusion. Since this study relies on a composite-sample scheme pairing public images with sampled environmental distributions, we developed a foundational proof-of-concept architecture (called CornSentry) rather than a field-validated biological study. Our architecture employs a late-fusion strategy which combines high-level visual representations extracted by Convolutional Neural Networks (CNNs) with in-situ environmental time series (temperature, humidity, soil moisture) modeled by Recurrent Neural Networks (RNNs). Furthermore, the framework integrates a Deep Convolutional Generative Adversarial Network (DCGAN) to overcome class imbalance and broaden feature diversity through synthetic data generation. We evaluated CornSentry using a comprehensive dataset combining visual and real-world environmental loT data. The results obtained demonstrate that our proposed multimodal late-fusion approach achieves a detection accuracy exceeding 98%, yielding a statistically significant improvement (p < 0.05) and a 3–4% increase over unimodal models that rely solely on image-based detection.</div></div>","PeriodicalId":29968,"journal":{"name":"Internet of Things","volume":"38 ","pages":"Article 102007"},"PeriodicalIF":7.1,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148499382","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

Adversarial attacks against network intrusion detection systems: Bridging the gap between theoretical vulnerabilities and practical constraints 针对网络入侵检测系统的对抗性攻击：弥合理论漏洞与实际约束之间的差距

IF 7.1 3区计算机科学

Internet of Things Pub Date : 2026-07-01 Epub Date: 2026-06-08 DOI: 10.1016/j.iot.2026.101997

Md Mehedi Hasan, Rafiqul Islam, Quazi Mamun, Md Zahidul Islam, Junbin Gao

{"title":"Adversarial attacks against network intrusion detection systems: Bridging the gap between theoretical vulnerabilities and practical constraints","authors":"Md Mehedi Hasan, Rafiqul Islam, Quazi Mamun, Md Zahidul Islam, Junbin Gao","doi":"10.1016/j.iot.2026.101997","DOIUrl":"10.1016/j.iot.2026.101997","url":null,"abstract":"<div><div>Network Intrusion Detection Systems (NIDS) have become critical components of cybersecurity infrastructure, with machine learning-based approaches achieving detection accuracies exceeding 99% on benchmark datasets. However, the emergence of adversarial machine learning has exposed fundamental vulnerabilities in these systems, where carefully crafted perturbations can evade detection with success rates surpassing 97% in laboratory conditions. This comprehensive literature review examines adversarial attacks against NIDS. Our investigation reveals a critical disconnect between theoretical attack capabilities and practical implementation constraints. Examining attack methodologies (gradient-based, GAN-based, black-box, and adaptive), the feature-space to network-packet translation challenge, defense mechanisms, and dataset limitations, this review applies a constraint-aware framework to quantify practical attack viability. Protocol compliance, semantic preservation, and infrastructure filtering render up to 80.3% of adversarial examples invalid under real-world constraints. Defense mechanisms including adversarial training, ensemble methods, and moving target defense face scalability barriers limiting production deployment. These findings suggest that many published attacks substantially overstate real-world vulnerabilities, providing a more nuanced understanding of practical threat landscapes for ML-based network security systems.</div></div>","PeriodicalId":29968,"journal":{"name":"Internet of Things","volume":"38 ","pages":"Article 101997"},"PeriodicalIF":7.1,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148523717","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

FedMP-LDL: Lightweight DDoS detection and mitigation in SDN-Fog IoT networks FedMP-LDL: SDN-Fog物联网中的轻量级DDoS检测和缓解

IF 7.1 3区计算机科学

Internet of Things Pub Date : 2026-07-01 Epub Date: 2026-06-08 DOI: 10.1016/j.iot.2026.101996

M.S. Gowtham, K.S. Bhuvaneshwari, M. Ramkumar, R. Sarath Kumar

{"title":"FedMP-LDL: Lightweight DDoS detection and mitigation in SDN-Fog IoT networks","authors":"M.S. Gowtham, K.S. Bhuvaneshwari, M. Ramkumar, R. Sarath Kumar","doi":"10.1016/j.iot.2026.101996","DOIUrl":"10.1016/j.iot.2026.101996","url":null,"abstract":"<div><div>Software-Defined Networking (SDN) integrated with fog-enabled internet of things environments improves scalability, flexibility, and centralized network management. However, the presence of heterogeneous IoT devices and resource-constrained fog nodes significantly increases vulnerability to large-scale Distributed Denial-of-Service (DDoS) attacks. Existing DDoS detection solutions often rely on centralized learning or computationally intensive deep learning models, which leads to higher latency and reduced real-time applicability. To address these issues, this research proposes a Federated Multi-Phase Lightweight Deep Learning (FedMP-LDL) framework for efficient DDoS detection and mitigation in SDN–fog–IoT networks. The novelty of FedMP-LDL is derived from its multi-phase architecture integrating Hamacher-based fuzzy equivalence rules for lightweight anomaly detection and a progressive distilled ghost-squeeze excitation network for hierarchical federated learning across fog nodes. In addition, multi-objective reinforcement learning is incorporated to enable adaptive SDN flow-rule optimization and dynamic DDoS mitigation in fog-enabled IoT environments. This distributed strategy eliminates dependency on centralized training, enhances scalability, and maintains robust performance under varying attack intensities. Experimental results demonstrate that FedMP-LDL achieves 99.72% accuracy, 99.66% precision, 99.57% recall, and 99.61% F1-score with a low inference time of 0.1762 s, outperforming existing state-of-the-art approaches while ensuring an effective balance between detection accuracy and computational efficiency.</div></div>","PeriodicalId":29968,"journal":{"name":"Internet of Things","volume":"38 ","pages":"Article 101996"},"PeriodicalIF":7.1,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148523719","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 of a PCB with microcontroller optimized for thermal control for IoT and industrial applications 针对物联网和工业应用的热控制优化的带有微控制器的PCB设计

IF 7.1 3区计算机科学

Internet of Things Pub Date : 2026-07-01 Epub Date: 2026-06-09 DOI: 10.1016/j.iot.2026.101995

Jesús Fernández-Fernández, Lidia Sánchez-González, Juan Felipe García-Sierra

{"title":"Design of a PCB with microcontroller optimized for thermal control for IoT and industrial applications","authors":"Jesús Fernández-Fernández, Lidia Sánchez-González, Juan Felipe García-Sierra","doi":"10.1016/j.iot.2026.101995","DOIUrl":"10.1016/j.iot.2026.101995","url":null,"abstract":"<div><div>This study presents the design and benefits of an open-source, optimized printed circuit board (PCB) controller, named Lionheart Pro, tailored for applications in thermal control, IoT, and industrial automation. By integrating an Atmel AT90USB1286 microcontroller, this PCB delivers enhanced connectivity, energy efficiency, and scalability, satisfying the diverse needs of projects such as 3D printing and automation. The research emphasizes the PCB’s hardware and software attributes, including compatibility with popular connectivity standards (WiFi, Bluetooth, USB), robust power management capabilities, and highly efficient thermal dissipation design, ensuring optimal temperature control and stability during high-performance operations. The versatility of the board was validated through two practical applications: as a central controller for an FDM 3D printer and as an automation driver for a GoTo telescope mount. The Lionheart Pro-PCB demonstrated consistent performance and reliability in these scenarios, underscoring its potential as a flexible and cost-effective solution for IoT-based systems and other digital control applications.</div></div>","PeriodicalId":29968,"journal":{"name":"Internet of Things","volume":"38 ","pages":"Article 101995"},"PeriodicalIF":7.1,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148523722","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

An intelligent cloud-based framework for healthcare cybersecurity using sparsity-aware edge attention network and sculptor optimization algorithm 基于云的医疗网络安全智能框架，使用稀疏感知边缘关注网络和雕刻家优化算法

IF 7.1 3区计算机科学

Internet of Things Pub Date : 2026-07-01 Epub Date: 2026-06-16 DOI: 10.1016/j.iot.2026.102003

G. Kalaiarasi, R. Poorni, Pradeep K․V․, Venkatraman K․

{"title":"An intelligent cloud-based framework for healthcare cybersecurity using sparsity-aware edge attention network and sculptor optimization algorithm","authors":"G. Kalaiarasi, R. Poorni, Pradeep K․V․, Venkatraman K․","doi":"10.1016/j.iot.2026.102003","DOIUrl":"10.1016/j.iot.2026.102003","url":null,"abstract":"<div><div>An intelligent cloud-based framework for healthcare cybersecurity offers a secure and scalable environment for managing patient data and protecting it from cyber threats. However, challenges such as privacy risks from third-party storage and potential service disruptions due to internet dependency remain. In order to solve such problems, a blockchain-driven Sparsity Aware Edge Attention Network with Sculptor Optimization Algorithm (SAEAN-SOA) scheme is presented in this research paper. Initially, the input data set obtained from the CICIoMT2024 and WUSTL EHMS 2020 dataset is pre-processed with the help of Memory-Boosted Guidance Filtering (MBGF). Subsequently, the filtered data set is subjected to feature selection through Meerkat Bipolar Swarm Optimization (MBSO) technique in order to select the best features. Such chosen features are then classified by Sparsity Aware Edge Attention Network (SAEAN) technique for detecting abnormal traffic flow. Lastly, an algorithm known as a blockchain-based Fair Consensus with Heterogeneous Miners (FCHM) is employed to provide safe and secured as well as tamper-proof sharing of medical data in the cloud. The overall SAEAN-SOA framework introduced in this paper provides a robust solution to the enhancement of security, scalability, and dependability of systems of healthcare integrated with cloud technologies. The experiment conducted on the performance of the suggested SAEAN-SOA shows promising results from two datasets, i.e., CICIoMT2024 and WUSTL EHMS 2020, with up to 99.50% precision, 99.60% recall rate, as well as high accuracy and F1-scores.</div></div>","PeriodicalId":29968,"journal":{"name":"Internet of Things","volume":"38 ","pages":"Article 102003"},"PeriodicalIF":7.1,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148499267","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