Internet of Things最新文献

{"title":"Neuromorphic solar edge AI for sustainable wildfire detection","authors":"Raúl Parada","doi":"10.1016/j.iot.2025.101862","DOIUrl":"10.1016/j.iot.2025.101862","url":null,"abstract":"<div><div>This paper presents a feasibility study of a solar-autonomous wildfire detection system using neuromorphic edge AI on fixed-wing drones. Through a comprehensive year-long simulation over Parc del Garraf (Catalonia), we evaluate three edge computing platforms, Raspberry Pi 4, Google Coral TPU, and BrainChip Akida, integrated into solar-optimized eBee X drones. Results show that the BrainChip Akida achieves 4200 patrol hrs per yr, nearly three times that of traditional CPU systems, while maintaining 87 % solar energy autonomy. The Google Coral TPU and Raspberry Pi 4 reach 66 % and 52 % autonomy, respectively. Fleet scaling analysis demonstrates that increasing drone count from one to eight reduces median wildfire detection time from 18 to 2.2 hrs, surpassing critical response thresholds. Seasonal analysis reveals Akida-based systems can operate fully on solar energy during summer and most of spring and fall, minimizing grid dependency. These findings establish neuromorphic computing as a foundational technology for sustainable, perpetual environmental monitoring within the Internet of Robotic Things (IoRT).</div></div>","PeriodicalId":29968,"journal":{"name":"Internet of Things","volume":"36 ","pages":"Article 101862"},"PeriodicalIF":7.6,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145885160","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}

{"title":"Classifying user-created passwords using machine learning and natural language processing techniques","authors":"Binh Le Thanh Thai,&nbsp;Tsubasa Takii,&nbsp;Hidema Tanaka","doi":"10.1016/j.iot.2025.101854","DOIUrl":"10.1016/j.iot.2025.101854","url":null,"abstract":"<div><div>Passwords are the dominant authentication method. However, evaluating the strength of user-created passwords remains a significant challenge due to the influence of various external factors, such as language, culture, and keyboard layout. In this paper, we address the problem of classifying user-created passwords into predefined groups, rather than directly evaluating their strength. First, we assess the performance of classifiers utilizing eight machine learning (ML) algorithms and four Natural Language Processing techniques to identify the optimal combination of ML algorithms and feature extraction methods. Through this experiment, we determine that the classifier combining Bag-of-Words and Logistic Regression is the most effective approach for classifying user-created passwords. Subsequently, we propose a hierarchical classification model to enhance the performance of this classifier. Experimental results demonstrate that the proposed model achieves accuracy of 97.81 % and recall of 99.66 % for weak passwords.</div></div>","PeriodicalId":29968,"journal":{"name":"Internet of Things","volume":"36 ","pages":"Article 101854"},"PeriodicalIF":7.6,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145791908","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}

{"title":"Stability analysis for fog computing via Lyapunov function","authors":"Shih-Yu Chang ,&nbsp;Mayank Kapadia ,&nbsp;Peishun Yan ,&nbsp;Wei Duan","doi":"10.1016/j.iot.2026.101871","DOIUrl":"10.1016/j.iot.2026.101871","url":null,"abstract":"<div><div>Fog computing is an emerging paradigm for the Internet of Things (IoT), where system stability directly impacts reliability, performance, and user experience. Existing stability models often ignore application-level service completion or fail to capture dynamic interactions among sensors and fog nodes. This study addresses these gaps by establishing necessary conditions for fog nodes to process application services within finite time. First, we introduce a fluid model based on a partial differential equation (PDE) to quantify the dynamics of service counts for each sensor when fog nodes are shared. Second, we design a Lyapunov function derived from the PDE solution to analyze system stability and convergence. Third, we apply this Lyapunov function to derive conditions that guarantee timely service completion. Finally, numerical experiments validate the fluid model, investigate PDE solution behavior, and assess the convergence speed of the Lyapunov function under various system parameters. These results provide actionable insights for ensuring stability and efficiency in fog computing systems for IoT applications.</div></div>","PeriodicalId":29968,"journal":{"name":"Internet of Things","volume":"36 ","pages":"Article 101871"},"PeriodicalIF":7.6,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146023231","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}

{"title":"FedMamba: Robust multimodal federated intrusion detection for heterogeneous IoT systems","authors":"Hafiz Bilal Ahmad ,&nbsp;Haichang Gao ,&nbsp;Naila Latif ,&nbsp;Tanjeena Manzoor","doi":"10.1016/j.iot.2026.101877","DOIUrl":"10.1016/j.iot.2026.101877","url":null,"abstract":"<div><div>The convergence of Information Technology (IT) and Operational Technology (OT) in Industry 4.0 produces diverse data streams, such as system logs, sensor readings, and network traffic, which are vital for industrial security. However, existing security analytics are siloed by modality and rely on centralized processing, raising concerns regarding privacy, latency, and scalability. Although Federated Learning (FL) mitigates privacy risks, most frameworks remain unimodal, lack support for non-IID data distributions, and face adversarial evasion challenges. We propose FedMamba, a novel multimodal Federated Learning (MMFL) framework that creates a unified Mamba-based model to address these issues via (i) efficient cross-modal learning, (ii) a FedProx-based protocol for stable non-IID training that remains compatible with secure aggregation, and (iii) modality-specific adversarial training for robustness. Experiments on HDFS, SWaT, and CICIoMT-2024 datasets show that the standard FedMamba achieved competitive macro F1-scores of 0.9584, 0.9795, and 0.9665 relative to centralized baselines, but degraded on HDFS and SWaT under PGD attack (0.3791 and 0.5147), whereas CICIoMT-2024 remained robust under the same attack (0.9665). The adversarially trained FedMamba-AT sustains robust F1-scores (0.9480, 0.8357, 0.9645). FedMamba offers a robust and scalable solution for secure IIoT monitoring.</div></div>","PeriodicalId":29968,"journal":{"name":"Internet of Things","volume":"36 ","pages":"Article 101877"},"PeriodicalIF":7.6,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146077994","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}

{"title":"A robust weighted late fusion approach for IoT","authors":"Vipin Nirala,&nbsp;Ratneshwer","doi":"10.1016/j.iot.2025.101857","DOIUrl":"10.1016/j.iot.2025.101857","url":null,"abstract":"<div><div>IoT systems consist of numerous sensors that generate multiple modalities. Consequently, IoT-based systems often rely on these multiple modalities to perform their intended functionalities. However, the dynamic, heterogeneous, and openly-deployed nature of IoT-based systems makes them susceptible to several issues, including data corruption, inconsistencies, and unavailability. Catastrophic impacts such as node failures or security vulnerabilities further hinder the consistent availability of modalities. Therefore, achieving reliable access to all modalities at all times is almost impossible. In this work, we aim to improve decision-making in IoT systems by adaptively weighing different modalities. To this end, we propose an adaptive weighted late fusion, which combines heuristic-based strategies with optimization-based weight adaptation. This hybrid approach maintains a balance between heuristics and optimization, thereby improving overall system performance. We compare the proposed work against popular multimodal fusion approaches, including accuracy-based, F1-based, and entropy-based weighted fusion methods. Experimental results show that our proposed fusion method outperforms these approaches in terms of raw performance. Additionally, we simulate scenarios involving data corruption and modality unavailability, in which our proposed fusion method demonstrates superior performance compared to benchmark methods. In conclusion, the proposed fusion approach performs better in both ideal scenarios and challenging conditions with modality unavailability and inconsistencies.</div></div>","PeriodicalId":29968,"journal":{"name":"Internet of Things","volume":"36 ","pages":"Article 101857"},"PeriodicalIF":7.6,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145841108","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}

{"title":"Correct by design, complete by iteration: A graph-based framework for automated security assessment","authors":"Felice Moretta ,&nbsp;Umberto Barbato ,&nbsp;Massimiliano Rak ,&nbsp;Daniele Granata","doi":"10.1016/j.iot.2025.101851","DOIUrl":"10.1016/j.iot.2025.101851","url":null,"abstract":"<div><div>Modern IT infrastructures, especially IoT and cyber-physical systems, require systematic and repeatable security assessment methods. A persistent challenge concerns the correctness and completeness of the system model underlying such analyses, which directly affect the quality of threat modeling and penetration test planning. Existing model-based approaches support these activities, but rarely ensure that the adopted model is both structurally valid and representative of the real system. This paper addresses this gap by introducing a formal modeling framework and extending a security assessment methodology (ESSecA) with a cyclical refinement process. At its core lies the Multi-purpose Application Composition Model (MACM), a property-graph formalism equipped with a schema and a set of syntactic and semantic constraints that define the space of admissible models. These constraints are automatically verified through formal checks (e.g., Cypher queries and Neo4j triggers), enabling automated verification of model correctness throughout the assessment lifecycle. As a result, any model accepted by the framework is guaranteed to comply with the rules of the modeling system. The cyclical refinement process complements this by addressing model completeness. Penetration testing results are iteratively reintegrated into the model, enriching it with newly discovered elements and interactions. This produces progressively more accurate system representations, which in turn yield more comprehensive threat models and increasingly precise penetration test plans, effectively mitigating grey-box limitations. The contribution is demonstrated through two case studies: the eWeLink IoT ecosystem, illustrating MACM’s modeling and validation capabilities, and the JetRacer autonomous vehicle platform, showcasing the full iterative methodology. Overall, the proposed approach combines a formal modeling system with a cyclic refinement process that exploits such formal guarantees to progressively enhance model completeness, ultimately strengthening threat modeling and penetration test planning.</div></div>","PeriodicalId":29968,"journal":{"name":"Internet of Things","volume":"36 ","pages":"Article 101851"},"PeriodicalIF":7.6,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145841107","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}

{"title":"HiFEL-OCKT: Hierarchical federated edge learning with objective congruence and multi-level knowledge transfer for IoT ecosystems","authors":"Ahmed-Rafik Baahmed,&nbsp;Jean-François Dollinger,&nbsp;Mohamed El Amine Brahmia,&nbsp;Mourad Zghal","doi":"10.1016/j.iot.2025.101868","DOIUrl":"10.1016/j.iot.2025.101868","url":null,"abstract":"<div><div>The explosive growth of Internet of Things (IoT) data and the demand for real-time decisions necessitate edge intelligence to overcome the latency and bandwidth limitations of cloud-only processing. Real-world IoT ecosystems are characterized by their high heterogeneity, which results from a wide variety of devices, sensors, environments, data, tasks, and resources, posing significant communication and computation efficiency challenges, scalability issues, and privacy concerns for edge intelligence. We propose HiFEL-OCKT, a novel hierarchical federated edge learning methodology for addressing the realistic high heterogeneity of IoT ecosystems, while enabling efficient edge intelligence. The key novelty of our proposed HiFEL-OCKT methodology is the efficient and scalable deployment of temporal intelligence at the edge by exploiting the valuable knowledge flowing at this level, which we define with the learning objective evolution, to ensure robust edge personalization through objective congruent collaboration and multi-level knowledge transfer between IoT devices. Through extensive experiments on multiple IoT domains, including smart buildings and industrial IoT with heterogeneous real-world datasets, our HiFEL-OCKT approach uncovered the novel ability in collaborating various highly heterogeneous IoT devices from different ecosystem settings. Our approach demonstrates superior performance and efficiency compared to the state-of-the-art approaches, with an improvement rate as high as 87.57 % in the edge knowledge personalization, while achieving significant speedups as high as 4.38 ×  in local training.</div></div>","PeriodicalId":29968,"journal":{"name":"Internet of Things","volume":"36 ","pages":"Article 101868"},"PeriodicalIF":7.6,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145926730","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}

{"title":"Geometry-aware and approximation-free Dijkstra optimization for discrete and continuous pareto fronts for task offloading in edge computing","authors":"Hesam Nejati SharifAldin,&nbsp;Farnoush NayebiPour","doi":"10.1016/j.iot.2026.101888","DOIUrl":"10.1016/j.iot.2026.101888","url":null,"abstract":"<div><div>Energy–performance optimization has become a central challenge in edge computing, where heterogeneous devices must decide whether to execute tasks locally or offload them to nearby servers. This bi-objective trade-off between execution time and energy consumption is difficult to resolve accurately: existing methods based on mixed-integer programming, evolutionary heuristics, or deep reinforcement learning often yield approximate Pareto fronts, require extensive tuning, and fail to scale under real-time constraints.</div><div>this paper propose a <em>geometry-aware optimization algorithm</em> that computes exact discrete and continuous Pareto fronts without any approximation or stochastic exploration. The method reformulates task-offloading decisions in a piecewise-linear execution–energy space and extends <em>Dijkstra’s</em> algorithm to propagate only non-dominated cost vectors according to Pareto dominance. This extension transforms the classical shortest-path search into a dual-objective process with provable guarantees of correctness and completeness. Theoretical analysis shows a bounded complexity of <span><math><mrow><mi>O</mi><mo>(</mo><mo>|</mo><mi>E</mi><mo>|</mo><mspace></mspace><mo>·</mo><mspace></mspace><mo>|</mo><mi>P</mi><msup><mo>|</mo><mn>2</mn></msup><mo>)</mo></mrow></math></span>, ensuring scalability even in dense edge networks.</div><div>Comprehensive experiments validate both accuracy and efficiency. Decision time grows near-linearly with problem size, while the average number of non-dominated vectors per node increases sub-linearly, confirming manageable computational growth. The constructed Pareto fronts remain stable under  ± 10% perturbations of input cost models and maintain consistent quality across increasing workloads, achieving the highest hypervolume (0.92) and lowest IGD (0.06) among all baselines. These results establish the proposed <em>extended-Dijkstra formulation</em> as a deterministic, interpretable, and scalable foundation for multi-objective task offloading in energy-constrained edge environments.</div></div>","PeriodicalId":29968,"journal":{"name":"Internet of Things","volume":"36 ","pages":"Article 101888"},"PeriodicalIF":7.6,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146173156","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}

{"title":"Real-time scalable UAV condition monitoring framework with hardware-level acceleration for IoT applications","authors":"Antonios Ntib ,&nbsp;Dimitrios Michael Manias ,&nbsp;Abdallah Shami","doi":"10.1016/j.iot.2025.101865","DOIUrl":"10.1016/j.iot.2025.101865","url":null,"abstract":"<div><div>Unmanned Aerial Vehicles (UAVs) are a critical component of emerging Smart Cities, supporting applications such as emergency response, transportation, environmental monitoring, and infrastructure inspection. Ensuring their reliability requires condition monitoring frameworks capable of real-time defect detection despite the UAVs’ resource-constrained nature. This work presents a real-time, scalable UAV condition monitoring framework that efficiently distributes computation between core servers and edge nodes. The novelty of this work lies in two main contributions: (i) the design and realization of a deployable framework tailored for real-time UAV monitoring with offloading capabilities to both edge and core resources, and (ii) the integration of hardware-level acceleration strategies, including OpenMP-based parallelization and AVX2 SIMD vectorization, to substantially enhance computational efficiency, scalability, and real-time feasibility. Together, these contributions position the framework as a practical solution ready for large-scale UAV swarm deployments. The overall improvements include significant reductions in processing time and enhanced resource utilization while maintaining predictive performance. A comparative evaluation across three frameworks, a baseline state-of-the-art Python framework, an intermediate C++/Cython translation, and the proposed fully optimized OpenMP/AVX2-based framework, demonstrates the framework’s readiness for integration into critical UAV-enabled IoT systems.</div></div>","PeriodicalId":29968,"journal":{"name":"Internet of Things","volume":"36 ","pages":"Article 101865"},"PeriodicalIF":7.6,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145885162","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}

{"title":"Snip-Cache: A code snippet caching system for LLM-based command-driven IoT systems","authors":"Chiwon Song ,&nbsp;Sooyong Kang","doi":"10.1016/j.iot.2025.101852","DOIUrl":"10.1016/j.iot.2025.101852","url":null,"abstract":"<div><div>Large language models (LLMs) are widely used in real-time interface systems that process user commands. Despite their high output quality, the long response times and substantial operating costs undermine the practicality and sustainability of LLM-based services. Prompt caching is one of the optimization techniques introduced to mitigate the problem. It avoids redundant processing of repetitive prompts by caching and reusing the response for the same or similar prompts. However, such a static caching scheme has an intrinsic limitation, in terms of the reusability of results, due to the variety of expressions having the same semantics in real-world usage environments. In this paper, we introduce a new strategy for prompt caching, <em>Snippet Caching</em>, for LLM-based command-driven IoT systems to overcome the limitation. It perceives a command (prompt) as a function call with specific arguments. Instead of caching (input, output) pairs, it caches two simple code snippets that mimic LLM operations for each function. Based on the strategy, we design a novel prompt caching scheme, <em>Snip-Cache</em>, which generates code snippets with the help of LLMs. Experimental results show that Snip-Cache is significantly more beneficial to command-driven IoT systems than semantic caching schemes (GPTCache and vCache), in terms of response accuracy, response time, and token usage.</div></div>","PeriodicalId":29968,"journal":{"name":"Internet of Things","volume":"36 ","pages":"Article 101852"},"PeriodicalIF":7.6,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145791860","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}