Internet of Things最新文献

筛选
英文 中文
Secrecy performance analysis of a multi-user industrial IoT network under multiple eavesdropper threats 多窃听者威胁下多用户工业物联网网络保密性能分析
IF 7.1 3区 计算机科学
Internet of Things Pub Date : 2026-11-01 Epub Date: 2026-10-26 DOI: 10.1016/j.iot.2026.102148
Jonathan Serugunda, Tonny Ssettumba, Abubaker Matovu Waswa, Patrick Odong
{"title":"Secrecy performance analysis of a multi-user industrial IoT network under multiple eavesdropper threats","authors":"Jonathan Serugunda,&nbsp;Tonny Ssettumba,&nbsp;Abubaker Matovu Waswa,&nbsp;Patrick Odong","doi":"10.1016/j.iot.2026.102148","DOIUrl":"10.1016/j.iot.2026.102148","url":null,"abstract":"<div><div>Industrial Internet of Things (IIoT) networks deployed in critical infrastructure are highly vulnerable to passive eavesdropping due to their open wireless medium, resource-constrained devices, and lack of centralized key management. Ensuring secure communication in such environments is still an open research challenge, particularly under multi-user and multi-eavesdropper settings where conventional cryptographic approaches may be impractical. In this paper, we propose a secure and efficient physical layer transmission scheme for IIoT environments. The system employs transmit antenna selection (TAS) at the base station and distributed selection combining (SC) over a set of cooperative single-antenna legitimate IIoT nodes connected to a controller, where the nodes act as spatially distributed receive branches for a common confidential transmission. A dedicated multi-antenna cooperative jammer applies zero-forcing beamforming (ZFB), forming the TAS-SC-ZFB scheme. Its beamforming vector suppresses artificial-jamming interference at the selected legitimate branch while degrading multiple eavesdroppers’ reception quality in the considered IIoT network. Such a design is especially suitable for industrial settings where devices are resource-constrained, channels are vulnerable, and centralized cryptographic management is impractical. We analyze the secrecy performance under both colluding and non-colluding eavesdropping scenarios and derive exact closed-form expressions for the secrecy outage probability (SOP), asymptotic SOP at high signal to noise ratio, and for the colluding case, we also analyze the ergodic secrecy rate. Monte Carlo simulations verify the analytical expressions. The proposed TAS-SC-ZFB scheme improves physical layer secrecy in the considered IIoT network by reducing secrecy outage probability and increasing effective and ergodic secrecy rates under the stated channel model and CSI assumptions.</div></div>","PeriodicalId":29968,"journal":{"name":"Internet of Things","volume":"40 ","pages":"Article 102148"},"PeriodicalIF":7.1,"publicationDate":"2026-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149354389","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 blockchain-based trust-weighted resilience mechanism for IoT networks 基于区块链的物联网网络信任加权弹性机制
IF 7.1 3区 计算机科学
Internet of Things Pub Date : 2026-11-01 Epub Date: 2026-11-02 DOI: 10.1016/j.iot.2026.102153
Alexandru A. Maftei, Alexandru Lavric, Adrian I. Petrariu, Eugen Coca, Valentin Popa
{"title":"A blockchain-based trust-weighted resilience mechanism for IoT networks","authors":"Alexandru A. Maftei,&nbsp;Alexandru Lavric,&nbsp;Adrian I. Petrariu,&nbsp;Eugen Coca,&nbsp;Valentin Popa","doi":"10.1016/j.iot.2026.102153","DOIUrl":"10.1016/j.iot.2026.102153","url":null,"abstract":"<div><div>The rapid development of decentralized IoT systems has made the device admission process increasingly difficult, especially in environments where there is no central authority that can prevent malicious devices from influencing distributed decisions in IoT networks. Currently, most blockchain-based solutions and trust mechanisms focus separately on either reputation evaluation or access control, without clearly defining how distributed entities can jointly evaluate, validate, and enforce admission decisions in dynamic IoT environments. Many admission processes rely on fixed thresholds for trust evaluation or implicit assumptions about device behavior, which can affect the scalability and consistency of decisions. In this paper, a decentralized admission mechanism for IoT devices based on blockchain is proposed, which combines multi-attribute evaluation, trust-weighted distributed voting, conditional event-aware verification, and efficient event validation in a single framework. Candidate IoT devices are evaluated using both reliability and performance information from trusted vendors and data obtained from their behavior during network interaction functions. Admission decisions are made collaboratively, through a voting mechanism in which the influence of each IoT device depends on its trust score. When suspicious behavior is detected, the conditional event-aware mechanism uses independently collected evidence as the reference for subsequent trust and weight updates, limiting the reinforcement of malicious evaluator influence. For efficient recording and validation of admission events, a Hierarchical Multi-level Merkle Reduction (HMMR) structure is introduced, providing lower memory usage and shorter construction time compared with the evaluated Merkle Tree and Merkle Mountain Range implementations, while exhibiting a trade-off in proof-generation time and proof size. The obtained results show that admission processing time ranges from 7.58 ms for 50 devices to 6.56 s for 20,000 devices. The proposed mechanism can efficiently combine IoT device evaluation, distributed decision-making, adaptive trust management, and validation into a single solution for managing device admission in large-scale decentralized IoT architectures. The evaluation was conducted in a controlled cloud-based emulation environment; therefore, hardware-specific resource consumption and heterogeneous real-world network conditions remain to be validated in future work.</div></div>","PeriodicalId":29968,"journal":{"name":"Internet of Things","volume":"40 ","pages":"Article 102153"},"PeriodicalIF":7.1,"publicationDate":"2026-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149354342","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
Trust-driven multi-criteria optimization framework for aerial MEC enabling resilient IoT services 用于空中MEC的信任驱动多标准优化框架,支持弹性物联网服务
IF 7.1 3区 计算机科学
Internet of Things Pub Date : 2026-11-01 Epub Date: 2026-10-30 DOI: 10.1016/j.iot.2026.102147
Muhammad Omair Butt, Muhammad Naeem, Waleed Ejaz
{"title":"Trust-driven multi-criteria optimization framework for aerial MEC enabling resilient IoT services","authors":"Muhammad Omair Butt,&nbsp;Muhammad Naeem,&nbsp;Waleed Ejaz","doi":"10.1016/j.iot.2026.102147","DOIUrl":"10.1016/j.iot.2026.102147","url":null,"abstract":"<div><div>Trustworthy edge computing increasingly depends on aerial platforms to extend connectivity and processing to disaster-affected and remote regions. High-altitude platforms (HAPs) and uncrewed aerial vehicles (UAVs) fulfill this role in emerging sixth-generation (6G) networks. UAVs’ resource constraints make their trustworthiness an essential input to the resource allocation decision. Existing works apply trust to a limited set of functions, such as stabilizing control policies or detecting threats, without connecting it to the allocation decision itself. This work develops a multi-dimensional trust model capturing reputation, stability, and security. This trust model enters the optimization as both a weighted objective term and a threshold constraint. We formulate a multi-criteria objective that jointly accounts for trust, deployment cost, trust-related cost, task latency, and IoT-UAV association. K-means clustering resolves UAV placement in the first stage. The proposed penalty-guided optimization (PGO) algorithm applies sequential quadratic programming and resolves trust-aware resource allocation in the second stage. PGO is evaluated against the benchmark algorithm, a greedy heuristic, and a relaxation-based baseline. PGO approaches the benchmark algorithm’s association quality across a wide range of network scales and trust requirements, at a fraction of its computational cost. These results demonstrate that trust-aware association and computational efficiency are jointly achievable. This establishes a practical basis for scalable deployment in future 6G aerial edge computing networks.</div></div>","PeriodicalId":29968,"journal":{"name":"Internet of Things","volume":"40 ","pages":"Article 102147"},"PeriodicalIF":7.1,"publicationDate":"2026-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149354386","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
Real-time AIoT platform-based bayesian optimization fine-tuning and deep sequence LSTM-CNN model for air quality prediction in smart buildings 基于实时AIoT平台的贝叶斯优化微调和深度序列LSTM-CNN模型在智能建筑空气质量预测中的应用
IF 7.1 3区 计算机科学
Internet of Things Pub Date : 2026-11-01 Epub Date: 2026-10-28 DOI: 10.1016/j.iot.2026.102152
Thanh-Phuong Nguyen
{"title":"Real-time AIoT platform-based bayesian optimization fine-tuning and deep sequence LSTM-CNN model for air quality prediction in smart buildings","authors":"Thanh-Phuong Nguyen","doi":"10.1016/j.iot.2026.102152","DOIUrl":"10.1016/j.iot.2026.102152","url":null,"abstract":"<div><div>Reliable forecasts of indoor air quality give the heating, ventilation, and air conditioning (HVAC) systems of smart buildings the lead time they need to act, which benefits both energy use and occupant comfort. This study proposes a real-time Artificial Intelligence of Things (AIoT) prediction framework in which Bayesian Optimization (BO) tunes a hybrid deep sequence model that chains a Long Short-Term Memory network with a Convolutional Neural Network (LSTM-CNN) to forecast several air quality variables at once. Five indicators are forecast together: temperature, humidity, CO2 concentration, fine particulate matter (PM2.5), and total volatile organic compounds (TVOCs), from high-frequency sensor streams gathered by a building-scale AIoT infrastructure. BO searches the hyperparameters of the LSTM-CNN automatically, and its search is weighed against Particle Swarm Optimization (PSO), a Genetic Algorithm (GA), and Random Search under the same budget. The tuned BO-LSTM-CNN is then compared with six common deep learning baselines, RNN, GRU, LSTM, Bi-RNN, BPNN, and CNN. Experimental results demonstrated that the proposed BO-LSTM-CNN improves prediction accuracy and generalization capability compared to all baseline methods. The results point to BO-tuned deep sequence learning on an AIoT sensing platform as a workable route to real-time air quality management and more energy-aware operation of smart buildings.</div></div>","PeriodicalId":29968,"journal":{"name":"Internet of Things","volume":"40 ","pages":"Article 102152"},"PeriodicalIF":7.1,"publicationDate":"2026-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149353924","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
Adaptive Load Balancing and Hybrid Scheduling for Energy-Efficient NB-IoT Operation in Integrated Terrestrial–Non-Terrestrial Networks 基于自适应负载均衡和混合调度的地非融合NB-IoT高能效运行
IF 7.1 3区 计算机科学
Internet of Things Pub Date : 2026-11-01 Epub Date: 2026-10-29 DOI: 10.1016/j.iot.2026.102151
Siamack Ghadimi, Thao Dinh Le, Martin Stusek, Dmitri Moltchanov, Pavel Masek, Jiri Misurec, Jiri Hosek
{"title":"Adaptive Load Balancing and Hybrid Scheduling for Energy-Efficient NB-IoT Operation in Integrated Terrestrial–Non-Terrestrial Networks","authors":"Siamack Ghadimi,&nbsp;Thao Dinh Le,&nbsp;Martin Stusek,&nbsp;Dmitri Moltchanov,&nbsp;Pavel Masek,&nbsp;Jiri Misurec,&nbsp;Jiri Hosek","doi":"10.1016/j.iot.2026.102151","DOIUrl":"10.1016/j.iot.2026.102151","url":null,"abstract":"&lt;div&gt;&lt;div&gt;Integrated terrestrial and non-terrestrial networks (TN–NTN) extend Narrowband Internet of Things (NB-IoT) coverage to remote regions but introduce intermittent connectivity and energy inefficiencies under partial satellite visibility. This paper presents a unified energy-aware framework that combines Adaptive TN–NTN Load Balancing (ATLB) with hybrid contention–grant uplink scheduling to optimize non-terrestrial access, improve network selection, and lower End Device (ED) energy consumption under intermittent Low Earth Orbit (LEO) NTN connectivity. The framework integrates battery-aware network assignment with a low-complexity per-packet scheduler that supports general energy–delay–reliability trade-offs; the evaluation reported here validates the energy-priority operating mode of this scheduler. We evaluate the performance of the proposed approach using 3GPP-aligned Monte Carlo simulations across multiple LEO constellations and ED densities (12–300 satellites, 50–5000 EDs), complemented by a two-dimensional Markov model that confirms the qualitative trends but is not used to validate numerical values. To comprehensively benchmark the proposed framework, we additionally formulate a Q-learning-based baseline and compare its performance against several representative schemes, including Adaptive Hybrid Scheduling and Access (AHSA), TN-only, NTN-only, and Threshold-based offloading. Our results show that ATLB keeps a flat mean delay across load, both at &lt;span&gt;&lt;math&gt;&lt;mrow&gt;&lt;mi&gt;M&lt;/mi&gt;&lt;mo&gt;=&lt;/mo&gt;&lt;mn&gt;12&lt;/mn&gt;&lt;/mrow&gt;&lt;/math&gt;&lt;/span&gt; ( ∼ 26 slots) and, up to &lt;span&gt;&lt;math&gt;&lt;mrow&gt;&lt;mi&gt;N&lt;/mi&gt;&lt;mo&gt;=&lt;/mo&gt;&lt;mn&gt;2000&lt;/mn&gt;&lt;/mrow&gt;&lt;/math&gt;&lt;/span&gt;, at &lt;span&gt;&lt;math&gt;&lt;mrow&gt;&lt;mi&gt;M&lt;/mi&gt;&lt;mo&gt;=&lt;/mo&gt;&lt;mn&gt;300&lt;/mn&gt;&lt;/mrow&gt;&lt;/math&gt;&lt;/span&gt; ( ∼ 8 slots). Per-device delay statistics reveal, however, that this flat mean masks a bimodal distribution of fast successes and attempts that exhaust the retry budget: at &lt;span&gt;&lt;math&gt;&lt;mrow&gt;&lt;mi&gt;M&lt;/mi&gt;&lt;mo&gt;=&lt;/mo&gt;&lt;mn&gt;12&lt;/mn&gt;&lt;/mrow&gt;&lt;/math&gt;&lt;/span&gt;, delay standard deviation is  ∼ 21 slots and the 95th/99th-percentile and maximum delay sit at the configured bound &lt;span&gt;&lt;math&gt;&lt;mrow&gt;&lt;msub&gt;&lt;mi&gt;D&lt;/mi&gt;&lt;mi&gt;max&lt;/mi&gt;&lt;/msub&gt;&lt;mo&gt;=&lt;/mo&gt;&lt;mn&gt;50&lt;/mn&gt;&lt;/mrow&gt;&lt;/math&gt;&lt;/span&gt; across the entire load range. AHSA does not show this behavior, maintaining near-zero delay standard deviation and a 95th-percentile/maximum delay of 8 slots over most of the evaluated range. AHSA therefore achieves lower average delay, energy, packet loss, and delay variability than ATLB across most operating points, with the gap widening under very high load. ATLB’s distinguishing behavior is instead a stable, load-independent reliance on NTN that avoids TN starvation, at the cost of higher delay, higher energy, and higher delay variability than AHSA. The primary contribution of this work is joint cross-layer integration of ATLB and Hybrid Contention–Grant Scheduling (HCGS) algorithms inducing battery- and load-aware network selection at the network layer coupled with energy-aware per-packet scheduling","PeriodicalId":29968,"journal":{"name":"Internet of Things","volume":"40 ","pages":"Article 102151"},"PeriodicalIF":7.1,"publicationDate":"2026-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149353925","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
The Cyprus Digital Twin architecture for resilient smart cities and critical infrastructure systems 塞浦路斯数字孪生体系结构,用于弹性智能城市和关键基础设施系统
IF 7.1 3区 计算机科学
Internet of Things Pub Date : 2026-11-01 Epub Date: 2026-10-27 DOI: 10.1016/j.iot.2026.102150
Philippos Isaia, Panayiotis Georgiou, Christophoros Leventis, Balaji Venkatasubramanian, Marios Kyriakou, Antonis Savva, Andreas Georgiou, Stelios Vrachimis, Christos Laoudias, Demetrios Eliades, Mathaios Panteli, Christos Panayiotou
{"title":"The Cyprus Digital Twin architecture for resilient smart cities and critical infrastructure systems","authors":"Philippos Isaia,&nbsp;Panayiotis Georgiou,&nbsp;Christophoros Leventis,&nbsp;Balaji Venkatasubramanian,&nbsp;Marios Kyriakou,&nbsp;Antonis Savva,&nbsp;Andreas Georgiou,&nbsp;Stelios Vrachimis,&nbsp;Christos Laoudias,&nbsp;Demetrios Eliades,&nbsp;Mathaios Panteli,&nbsp;Christos Panayiotou","doi":"10.1016/j.iot.2026.102150","DOIUrl":"10.1016/j.iot.2026.102150","url":null,"abstract":"<div><div>Digital twins are increasingly used to support the monitoring and management of urban infrastructure, yet their application to disaster preparedness and critical-infrastructure resilience remains fragmented across individual hazards and infrastructure sectors. This article presents the Cyprus Digital Twin (CyDT), a modular, scenario-driven platform for integrating heterogeneous data, infrastructure models, simulation tools, and stakeholder-oriented services within a common digital environment. CyDT combines geospatial, historical, demographic, and streaming data with distributed workflow execution, domain-specific simulators, interactive visualization, application programming interfaces, and controlled access to public and confidential datasets. The platform is demonstrated through four scenarios addressing wastewater contamination of a drinking-water network, blast propagation in an industrial area, wildfire interaction with power infrastructure, and windstorm impacts on co-located power and telecommunication assets. The scenarios show how CyDT connects hazard evolution with infrastructure exposure, potential failure, population impact, and alternative mitigation actions. In particular, the contamination scenario quantifies how network intervention can reduce population exposure, while the windstorm scenario demonstrates that infrastructure sectors exposed to the same hazard may experience significantly different disruption patterns because of differences in asset distribution and fragility. Collectively, the results illustrate the value of a reusable digital-twin architecture for exploring vulnerabilities, comparing response strategies, and communicating the spatial and temporal consequences of disruptive events. The study also identifies model validation, computational responsiveness, interoperability, data governance, and organizational integration as key requirements for progressing from research-oriented scenario analysis toward operational decision support.</div></div>","PeriodicalId":29968,"journal":{"name":"Internet of Things","volume":"40 ","pages":"Article 102150"},"PeriodicalIF":7.1,"publicationDate":"2026-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149353926","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
Symbolic representations for human activity recognition: From raw signals to lightweight sensor-language models 人类活动识别的符号表示:从原始信号到轻量级传感器语言模型
IF 7.1 3区 计算机科学
Internet of Things Pub Date : 2026-11-01 Epub Date: 2026-10-18 DOI: 10.1016/j.iot.2026.102142
Javier Medina-Quero, Jose Manuel Soto-Hidalgo, Oresti Ba&nos-Legrán, Aurora Polo-Rodríguez
{"title":"Symbolic representations for human activity recognition: From raw signals to lightweight sensor-language models","authors":"Javier Medina-Quero,&nbsp;Jose Manuel Soto-Hidalgo,&nbsp;Oresti Ba&nos-Legrán,&nbsp;Aurora Polo-Rodríguez","doi":"10.1016/j.iot.2026.102142","DOIUrl":"10.1016/j.iot.2026.102142","url":null,"abstract":"<div><div>Human Activity Recognition (HAR) commonly relies on raw-signal temporal modelling, often achieving strong performance at the expense of interpretability, efficiency, and semantic transparency. This work proposes a symbolic HAR framework in which statistical descriptors and fuzzy logic-based abstractions are used to represent sensor behaviour in a more interpretable and semantically meaningful form. These symbolic representations are exploited both as discriminative features for HAR models and as sensor/class signatures that support explanation generation in lightweight language models. Building on this representation, we introduce <em>Symbolic Sensor TinyLLaMA</em>, a HAR-oriented lightweight sensor-language framework designed for comparatively resource-efficient deployment. Rather than simply replacing a larger sensor-language model with a smaller one, the proposed approach reformulates sensor-language modelling around HAR-specific symbolic signatures, teacher-guided explanation generation, and dual implicit-explicit evaluation. In particular, tree-based teacher models, such as XGBoost, are used to guide the association between activity classes and discriminative symbolic features through a distillation-like strategy. The framework is evaluated on heterogeneous wearable and ambient-sensing datasets, including IMU-based and multi-occupancy smart-home environments. Experimental results compare raw-signal and symbolic representations across classical machine learning methods, deep learning architectures, boosted tree-based models, and sensor-aligned lightweight language models. The results show that symbolic representations can achieve competitive HAR performance while improving interpretability and enabling explicit textual explanations of sensor-based activity predictions.</div></div>","PeriodicalId":29968,"journal":{"name":"Internet of Things","volume":"40 ","pages":"Article 102142"},"PeriodicalIF":7.1,"publicationDate":"2026-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149354343","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 cognitive IoT currency framework for human-centered admission control via reinforcement learning 通过强化学习实现以人为中心的准入控制的认知物联网货币框架
IF 7.1 3区 计算机科学
Internet of Things Pub Date : 2026-11-01 Epub Date: 2026-10-23 DOI: 10.1016/j.iot.2026.102146
Yousry AbdulAzeem, Aboul Ella Hassanien
{"title":"A cognitive IoT currency framework for human-centered admission control via reinforcement learning","authors":"Yousry AbdulAzeem,&nbsp;Aboul Ella Hassanien","doi":"10.1016/j.iot.2026.102146","DOIUrl":"10.1016/j.iot.2026.102146","url":null,"abstract":"<div><div>The increasing use of Internet of Things devices in daily life creates challenges for human cognitive capacity, as they demand significant attention and can reduce user performance and well-being. Current IoT systems focus on technical efficiency without considering human cognitive states. This paper proposes a new framework for measuring and managing the mental cost of interacting with IoT devices using cognitive currency, which consists of Attention Tokens, Focus Credits, and Cognitive Load Units. A reinforcement learning engine uses Deep Q-Network (DQN) and tabular Q-learning to allow, delay, aggregate, or suppress requests to balance task performance and cognitive well-being. The framework is evaluated through synthetic simulation and a real dataset using ten training seeds. Without cognitive-load recovery, DQN reduces soft-overload steps by 100% in simulation and 99.5% on real data. The digital-twin overload penalty enables the learned policy to deliver fewer total requests than the unprotected baseline (7.0 vs 9.7 and 7.7 vs 12.5 per episode) but about 1.7–2.3 times the high-priority share. Threshold-based heuristics eliminate overload but deliver only 1–4 requests per episode, while DQN delivers 4–6 more with some overload. With load and attention recovery, an event-aware aggregation heuristic competes strongly with DQN, whose reward varies across settings. A queue extension shows that recovery encourages DQN to use delay and aggregation, although it delivers fewer requests than in the memoryless setting.</div></div>","PeriodicalId":29968,"journal":{"name":"Internet of Things","volume":"40 ","pages":"Article 102146"},"PeriodicalIF":7.1,"publicationDate":"2026-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149353923","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
Corrigendum to ‘An explainable neuro-fuzzy intrusion detection system for industrial IoT’ [Internet of Things 38 (2026)101989] “工业物联网可解释的神经模糊入侵检测系统”的勘误表[物联网38 (2026)101989]
IF 7.1 3区 计算机科学
Internet of Things Pub Date : 2026-09-01 Epub Date: 2026-06-20 DOI: 10.1016/j.iot.2026.101999
Malena Pérez Sevilla, José Ángel Olivas Varela, Álvaro Herrero Cosio, Daniel Urda Muñoz, Jaime Andrés Rincón Arango
{"title":"Corrigendum to ‘An explainable neuro-fuzzy intrusion detection system for industrial IoT’ [Internet of Things 38 (2026)101989]","authors":"Malena Pérez Sevilla,&nbsp;José Ángel Olivas Varela,&nbsp;Álvaro Herrero Cosio,&nbsp;Daniel Urda Muñoz,&nbsp;Jaime Andrés Rincón Arango","doi":"10.1016/j.iot.2026.101999","DOIUrl":"10.1016/j.iot.2026.101999","url":null,"abstract":"","PeriodicalId":29968,"journal":{"name":"Internet of Things","volume":"39 ","pages":"Article 101999"},"PeriodicalIF":7.1,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148940806","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
Corrigendum to “μTEE: Certification-ready RISC-V platform for secure IIoT lifecycle management” [Internet of Things 36 (2026) 101885] “μTEE:用于安全IIoT生命周期管理的认证就绪RISC-V平台”的勘误表[Internet of Things 36 (2026) 101885]
IF 7.1 3区 计算机科学
Internet of Things Pub Date : 2026-09-01 Epub Date: 2026-07-02 DOI: 10.1016/j.iot.2026.102018
Alessandro Cilardo, Jesús García-Rodríguez, Jose Manuel Merlos Espín, Roberto Nardone
{"title":"Corrigendum to “μTEE: Certification-ready RISC-V platform for secure IIoT lifecycle management” [Internet of Things 36 (2026) 101885]","authors":"Alessandro Cilardo,&nbsp;Jesús García-Rodríguez,&nbsp;Jose Manuel Merlos Espín,&nbsp;Roberto Nardone","doi":"10.1016/j.iot.2026.102018","DOIUrl":"10.1016/j.iot.2026.102018","url":null,"abstract":"","PeriodicalId":29968,"journal":{"name":"Internet of Things","volume":"39 ","pages":"Article 102018"},"PeriodicalIF":7.1,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148940805","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
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
相关产品
×
本文献相关产品
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:604180095
Book学术官方微信
小红书