Internet of Things最新文献

{"title":"TinyResViT: A lightweight hybrid deep learning model for on-device corn leaf disease detection","authors":"Van-Linh Truong-Dang,&nbsp;Huy-Tan Thai,&nbsp;Kim-Hung Le","doi":"10.1016/j.iot.2025.101495","DOIUrl":"10.1016/j.iot.2025.101495","url":null,"abstract":"<div><div>The increasing prevalence of corn leaf diseases poses a significant threat to global food security, necessitating efficient and accurate detection methods. To address this challenge, we introduce TinyResViT, a lightweight yet efficient hybrid deep learning model designed by combining Residual Network (ResNet) and Vision Transformer (ViT) for leaf disease detection. This combination leverages the strengths of ResNet in extracting local features and ViT in capturing global interactions among features. In addition, a novel downsampling block connecting ResNet and ViT is proposed to eliminate redundant model weights. The evaluation results on the PlantVillage and Bangladeshi Crops Disease datasets show TinyResViT’s superior performance, achieving F1-scores of 97.92% and 99.11%, respectively. The model also maintains a high processing speed of 83.19 Frames Per Second (FPS) and a low computational cost of 1.59 Giga Floating Point Operations (GFLOPs), outperforming existing deep neural networks and state-of-the-art approaches.</div></div>","PeriodicalId":29968,"journal":{"name":"Internet of Things","volume":"30 ","pages":"Article 101495"},"PeriodicalIF":6.0,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143222781","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":"Energy efficient resource management for real-time IoT applications","authors":"Rolden John Fereira ,&nbsp;Chathurika Ranaweera ,&nbsp;Kevin Lee ,&nbsp;Jean-Guy Schneider","doi":"10.1016/j.iot.2025.101515","DOIUrl":"10.1016/j.iot.2025.101515","url":null,"abstract":"<div><div>The Internet of Things (IoT) has a large and rapidly expanding number of deployed devices, which leads to a significant global energy consumption footprint. Diverse IoT use cases, including smart cities, smart grids, Industry 5.0, eHealth, and autonomous vehicles, are contributing to this increase in energy consumption. Optimising energy utilisation is crucial to sustaining the exponential growth of IoT applications, which demand stringent delays and latencies measured in milliseconds and microseconds. There are additional complexities with the emergence of edge, fog, and cloud computing and the need to manage the energy consumption at all the layers. In this paper, mechanisms that can be used to minimise energy consumption within an edge–fog–cloud IoT architecture for real-time IoT applications are being proposed. We investigate mechanisms for optimal node selection, primarily focusing on minimising energy usage while adhering to the Quality of Service (QoS) requirements of various IoT requests. The mechanisms include genetic, modified genetic, and delay-aware algorithms tailored explicitly for real-time IoT applications. We evaluated the proposed mechanisms using a simulation of diverse network scenarios. The results presented in the paper provide insight into balancing processing time and energy efficiency, which are critical considerations in sustainably developing IoT applications in an edge–fog–cloud IoT architecture.</div></div>","PeriodicalId":29968,"journal":{"name":"Internet of Things","volume":"30 ","pages":"Article 101515"},"PeriodicalIF":6.0,"publicationDate":"2025-01-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143348009","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}

{"title":"Automatic software tailoring for Green Internet of Things","authors":"José Miguel Aragón-Jurado,&nbsp;Juan Carlos de la Torre,&nbsp;Patricia Ruiz,&nbsp;Bernabé Dorronsoro","doi":"10.1016/j.iot.2025.101521","DOIUrl":"10.1016/j.iot.2025.101521","url":null,"abstract":"<div><div>The proliferation of low-capacity, interconnected Internet of Things devices has increased the need for energy efficient software. Optimizing software performance for specific hardware requires tailored code transformations, as universal compiler optimizations are insufficient. Moreover, the diversity of devices and the software running on them requires automating this process. This work presents a novel combinatorial optimization problem focused on minimizing software energy consumption for specific hardware, and a methodology for solving it that accounts for system uncertainty. Additionally, a novel device for measuring the energy consumption during run time is introduced. This meter synchronizes with the experiments, enabling the automatic optimization of software as a combinatorial optimization problem. Specifically, the problem involves finding an LLVM transformation sequence that minimizes energy consumption during software execution. In our experiments, we considered two different software benchmarks and two embedded devices, using a cellular genetic algorithm to optimize them, alongside five state-of-the-art approaches to manage uncertainty. Our results demonstrate that the proposed methodology successfully overcomes uncertainty, leading to greener solutions with improvements of up to 62.29% in run time and up to 58.21% in energy consumption, outperforming the best generic compilation flags by up to 32.12% in run time and 27.84% in energy consumption.</div></div>","PeriodicalId":29968,"journal":{"name":"Internet of Things","volume":"30 ","pages":"Article 101521"},"PeriodicalIF":6.0,"publicationDate":"2025-01-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143223142","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":"LWSEE: Lightweight Secured Software-Based Execution Environment","authors":"José Cecílio ,&nbsp;Alan Oliveira de Sá ,&nbsp;Georg Jäger ,&nbsp;André Souto ,&nbsp;António Casimiro","doi":"10.1016/j.iot.2025.101513","DOIUrl":"10.1016/j.iot.2025.101513","url":null,"abstract":"<div><div>The Internet of Things (IoT) has become increasingly prevalent and used to handle sensitive and critical data. This demands mechanisms to ensure data security, protect privacy, and promote the general safety of IoT-based systems. Currently, hardware-based trusted execution environments (TEEs) are used to provide data protection, but they are not suitable for low-cost devices lacking hardware-assisted security features. To address this issue, this paper proposes a Lightweight Secured Software-based Execution Environment (LWSEE) for embedded devices. LWSEE is designed to be supported by low-cost, low-end devices without specific hardware requirements. It consists of a lightweight distributed solution that offers protection against hardware attacks, provides a comprehensive security check mechanism, enables secure application execution, and supports secure application updates to ensure the continued security of IoT devices. LWSEE comprises a secure architecture and communication protocol specially tailored to devices with constrained resources. Our experimental evaluation underlines the minimal overhead introduced by LWSEE while showing its performance in terms of execution time, CPU time, and memory usage. We examine the flexibility and adaptability of LWSEE by demonstrating that it can be configured to achieve minimal overhead (<em>e.g.</em>, <span><math><mrow><mn>39</mn><mo>.</mo><mn>8</mn></mrow></math></span> ms per message for the general integrity verification of a node). This approach enables IoT devices to remain secure without dedicated hardware, allowing for the widespread adoption of IoT technology while maintaining data safety.</div></div>","PeriodicalId":29968,"journal":{"name":"Internet of Things","volume":"30 ","pages":"Article 101513"},"PeriodicalIF":6.0,"publicationDate":"2025-01-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143223144","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}

{"title":"Energy-aware tinyML model selection on zero energy devices","authors":"Adnan Sabovic ,&nbsp;Jaron Fontaine ,&nbsp;Eli De Poorter ,&nbsp;Jeroen Famaey","doi":"10.1016/j.iot.2025.101488","DOIUrl":"10.1016/j.iot.2025.101488","url":null,"abstract":"<div><div>Tiny Machine Learning (tinyML) enables the efficient deployment of machine learning models on resource-constrained Internet of Things (IoT) devices. However, in scenarios where energy availability is variable and unpredictable, such as with zero energy devices (ZEDs) reliant on energy harvesters, deployed tinyML models are often compressed to accommodate worst-case energy constraints, leading to diminished accuracy. To address this challenge, we propose a strategy of deploying multiple tinyML models concurrently on ZEDs to maximize model accuracy within the time-varying constraints of memory, execution time, and energy. We introduce a mathematical optimization framework that dynamically selects the most suitable tinyML model for execution based on current and predicted energy availability. We validate our approach through experimental evaluation, where we develop, train, and assess two machine learning models in a Cloud environment before optimizing them into tinyML models of differing sizes and accuracies. Our methodology is tested using a prototype for photovoltaic-powered battery-less gesture detection and recognition, employing a controlled setup with artificial lighting conditions. Results indicate that, under constant harvesting current, the smaller tinyML model exhibits superior execution speed, with two more executions on average, while the larger model yields approximately 28% higher accuracy. As expected, in a more realistic scenario with dynamic harvesting currents and employing our optimization algorithm, the device automatically prioritizes the larger tinyML model for inference when plentiful energy can be harvested due to its improved accuracy, while switching to the smaller model otherwise.</div></div>","PeriodicalId":29968,"journal":{"name":"Internet of Things","volume":"30 ","pages":"Article 101488"},"PeriodicalIF":6.0,"publicationDate":"2025-01-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143223145","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":"Machine learning-based optimal data retrieval and resource allocation scheme for edge mesh coupled information-centric IoT networks and disability support systems","authors":"Wilayat Khan ,&nbsp;Bilal Hassan ,&nbsp;Ramsha Ahmed ,&nbsp;Muhammad Nasir Bhutta ,&nbsp;Jawad Yousaf ,&nbsp;Kais Belwafi ,&nbsp;Mohamed Jleli ,&nbsp;Bessem Samet ,&nbsp;Taimur Hassan","doi":"10.1016/j.iot.2025.101511","DOIUrl":"10.1016/j.iot.2025.101511","url":null,"abstract":"<div><div>Cloud-centric computing, due to its lack of mobility and increased latency, is not suitable for addressing unprecedented challenges within an Internet of Things (IoT) network, especially in the context of disability support systems. However, recent advancements in edge computing provided an alternative to cloud servers by deploying the data processing tasks at the edge level, increasing both the efficiency and throughput of the IoT networks. This paper introduces a novel architecture, dubbed ICN-EdgeMesh, that fuses information-centric networking (ICN) with edge mesh computing to provide optimal data access within an IoT network. Furthermore, we employ Support Vector Machines (SVM) classification models to establish the edge-to-things continuum by allocating the optimal node to each IoT device within the network for retrieving the requested data. Moreover, we evaluate the performance of ICN-EdgeMesh against multiple key factors, where it achieved a high data rate (of 9.1 to 10 Mbps) along with ultra-low latency. In addition, the trained SVM model within the proposed scheme achieved 98.1% accuracy, with a true positive rate of 95.3% and a true negative rate of 98.8%, reflecting the optimal network node allocation for efficient data transmission.</div></div>","PeriodicalId":29968,"journal":{"name":"Internet of Things","volume":"30 ","pages":"Article 101511"},"PeriodicalIF":6.0,"publicationDate":"2025-01-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143222782","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":"Context-aware IoT search engine through fuzzy clustering: Search space restructuring and query resolution mechanisms","authors":"Santosh Pattar,&nbsp;Veena Badiger,&nbsp;Yash Kangralkar","doi":"10.1016/j.iot.2025.101494","DOIUrl":"10.1016/j.iot.2025.101494","url":null,"abstract":"<div><div>As the technological landscape of ubiquitous computing continues to develop and increase in the variety of objects being connected to sensors, a number of smart applications and services are being evolved that offer consumer-centric functionalities and solutions across different sectors. Therefore the search for sensors with capacity to provide most meaningful facts becomes paramount. Leveraging the inherent contextual metadata associated with sensor deployments and user applications can facilitate the systematic identification and removal of redundant or obsolete sensor infrastructure within large-scale organizational environments. This work proposes a novel approach that incorporates the principles of fuzzy context oriented seek algorithm that maps the submitted user question to the maximum appropriate sensors available within the search space. We have used the weighted fuzzy c-way cluster set of rules to institution sensors of similar houses into one cluster. So, the consumer query is channeled to the most relevant cluster. We have performed experiments and as compared the end result with the existing search algorithm noted inside the literature to uplift its performance.</div></div>","PeriodicalId":29968,"journal":{"name":"Internet of Things","volume":"30 ","pages":"Article 101494"},"PeriodicalIF":6.0,"publicationDate":"2025-01-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143223141","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":"MultiCIDS: Anomaly-based collective intrusion detection by deep learning on IoT/CPS multivariate time series","authors":"Marta Catillo,&nbsp;Antonio Pecchia,&nbsp;Umberto Villano","doi":"10.1016/j.iot.2025.101519","DOIUrl":"10.1016/j.iot.2025.101519","url":null,"abstract":"<div><div>Intrusion detection plays a key role to support secure operations of critical assets and services based on the Internet of Things (IoT) and cyber–physical systems. Most papers on the topic tend to favor the use of point anomaly approaches to detect intrusions by means of machine and deep learning. However, addressing intrusions through point anomaly approaches causes a major under-utilization of the monitoring data available. Differently from existing work, this paper proposes MultiCIDS, a novel approach that handles monitoring data as multivariate time series – typically available in any IoT system – to detect collective intrusions.</div><div>MultiCIDS capitalizes on a hybrid strategy, which pipelines a per-point scoring function implemented by a semi-supervised autoencoder and a sliding window algorithm. The evaluation is based on normal and intrusion time series pertaining to IoT devices, a cyber–physical system and a ubiquitous server. The benchmark datasets used in the experiment cover a wide spectrum of intrusions. The results indicate that MultiCIDS is competitive with other state-of-the-art deep learning techniques for handling sequential data. More importantly, MultiCIDS is characterized by negligible training–detection duration and achieves a major reduction of the false positives, which makes it suitable for real-life operations.</div></div>","PeriodicalId":29968,"journal":{"name":"Internet of Things","volume":"30 ","pages":"Article 101519"},"PeriodicalIF":6.0,"publicationDate":"2025-01-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143222793","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}

{"title":"A contactless method for recognition of daily living activities for older adults based on ambient assisted living technology","authors":"Kang Wang ,&nbsp;Ashish Saragadam ,&nbsp;Jasleen Kaur ,&nbsp;Ayan Dogra ,&nbsp;Shi Cao ,&nbsp;Moojan Ghafurian ,&nbsp;Zahid A. Butt ,&nbsp;Shahabeddin Abhari ,&nbsp;Dmytro Chumachenko ,&nbsp;Plinio P. Morita","doi":"10.1016/j.iot.2025.101502","DOIUrl":"10.1016/j.iot.2025.101502","url":null,"abstract":"<div><h3>Background</h3><div>During demographic shifts towards an older population, healthcare systems face increased demands, highlighting the need for innovative approaches that facilitate supporting older adults’ well-being and safety. This study aims to demonstrate the effectiveness of zero-effort Ambient Assisted Living technology in recognizing daily activities of older adults via machine learning algorithms by comparing with wearable technology.</div></div><div><h3>Methods</h3><div>Conducted in a smart home environment equipped with a comprehensive range of non-intrusive sensors, the study involved 40 participants, during which they were instructed to perform 23 types of predefined daily living activities, organized in five phases. Data from these activities were concurrently captured by both ambient and wearable sensors. Analysis was performed using five machine learning models: K-Nearest Neighbors, Decision Trees, Random Forest, Adaptive Boosting, and Gaussian Naive Bayes.</div></div><div><h3>Results</h3><div>Ambient sensors, especially using the AdaBoost model, demonstrated high accuracy (0.964) in activity recognition, significantly outperforming wearable sensors (best accuracy 0.367 with Random Forest). When fusing data from both sensor types, the accuracy slightly decreases to 0.909. Despite spatial overlap challenges, ambient sensors accurately recognize activities across various room settings with accuracies all above 0.950. Feature importance analysis reveals that climatic, electrical, and motion-related features are crucial for model classification.</div></div><div><h3>Conclusion</h3><div>This study showcases the efficacy of Ambient Assisted Living technology in recognizing daily indoor activities of older adults. These findings have implications for public health, highlighting Ambient Assisted Living technology's potential to support older adults' independence and well-being, offering a promising direction for future research and application in smart living environments.</div></div>","PeriodicalId":29968,"journal":{"name":"Internet of Things","volume":"30 ","pages":"Article 101502"},"PeriodicalIF":6.0,"publicationDate":"2025-01-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143223143","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}

{"title":"Lightweight privacy-protection RFID protocol for IoT environment","authors":"Kuo-Yu Tsai,&nbsp;You-Lin Wei,&nbsp;Po-Shen Chi","doi":"10.1016/j.iot.2025.101490","DOIUrl":"10.1016/j.iot.2025.101490","url":null,"abstract":"<div><div>The rapid growth of the Internet of Things (IoT for short) has expanded its applications across diverse domains, including smart healthcare, smart homes, and smart factories. Among the key technologies driving this evolution, Radio Frequency Identification (RFID for short) plays a pivotal role in IoT ecosystems due to its automation, identity recognition, and portability attributes. These features make RFID essential for simplifying device management and enhancing traceability in practical scenarios, particularly in healthcare, where it optimizes the management of patient medical records. However, frequent information exchanges within RFID systems pose a significant challenge, as inadequate authentication mechanisms can lead to unintended exposure of sensitive personal data. Fan <em>et al</em>. propose a lightweight RFID authentication protocol in IEEE Transactions on Industrial Informatics to address this issue. Unfortunately, our analysis finds several security vulnerabilities in their protocol, including susceptibility to impersonation, traceability, and secret disclosure attacks. In this paper, we develop a new lightweight privacy-protection RFID protocol, building upon Fan <em>et al</em>.’s framework. Our security evaluation demonstrates that the proposed protocol effectively mitigates these threats, ensuring the confidentiality and integrity of sensitive data in RFID-enabled systems.</div></div>","PeriodicalId":29968,"journal":{"name":"Internet of Things","volume":"30 ","pages":"Article 101490"},"PeriodicalIF":6.0,"publicationDate":"2025-01-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143222783","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}