Internet of Things最新文献

{"title":"An open testbed for O-RAN experimentation with AI-enabled control and monitoring","authors":"Raúl Parada ,&nbsp;Xavier Vilajosana ,&nbsp;Sobhi Alfayoumi ,&nbsp;Jordi Serra ,&nbsp;Oriol Font-Bach ,&nbsp;Paolo Dini","doi":"10.1016/j.iot.2025.101729","DOIUrl":"10.1016/j.iot.2025.101729","url":null,"abstract":"<div><div>The proliferation of open radio access networks (O-RAN) in modern 5G systems has ushered in enhanced flexibility and efficiency, but it has also introduced novel security challenges. In response, this paper presents a novel AI-based anomaly detection framework tailored for O-RAN networks operating in 5G environments. By employing principal component analysis for dimensionality reduction and a deep neural network for classification, the proposed system efficiently processes large-scale 5G traffic data while achieving high detection accuracy and low latency. Experimental evaluation on an open-source testbed with realistic cellular traffic demonstrates rapid convergence, with both training and validation accuracy values approaching 100% and effective detection of anomalies introduced via user equipment identifier swaps. The testbed processed over 300,000 traffic samples with 31 distinct network features, emulating 8 unique user equipment profiles under diverse radio conditions. Under adversarial scenarios, such as identity-swapping attacks, the system identified anomalous behavior with detection rates exceeding 40%, while maintaining a near-zero false positive rate on clean traffic. These results underscore the testbed’s capability to simulate complex 5G environments and the framework’s ability to deliver highly accurate, low-latency, and scalable anomaly detection. Overall, this work highlights the potential of advanced AI techniques to significantly enhance the security and resilience of modern wireless communication networks.</div></div>","PeriodicalId":29968,"journal":{"name":"Internet of Things","volume":"33 ","pages":"Article 101729"},"PeriodicalIF":7.6,"publicationDate":"2025-08-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144890107","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 simulation framework for assessing and optimizing IoT service and resource allocation: SimulateIoT-Services","authors":"Fernando Díaz Cantero,&nbsp;José Ángel Barriga Corchero,&nbsp;Miguel Ángel Pérez-Toledano,&nbsp;Pedro J. Clemente","doi":"10.1016/j.iot.2025.101736","DOIUrl":"10.1016/j.iot.2025.101736","url":null,"abstract":"<div><div>The Internet of Things (IoT) is increasingly being applied across various domains, including smart cities, smart buildings, agriculture, and connected vehicles, also known as the Internet of Vehicles (IoV), as well as industry, where it is commonly referred to as the Industrial IoT (IIoT). The ultimate goal of these systems is to provide services to end users. Services can be deployed across the computing layers of IoT systems, such as the fog or cloud layers, in a process known as service allocation. Similarly, hardware resources can be configured for each node within these layers, a process referred to as resource allocation. Properly executing these processes is crucial to achieving optimal performance in IoT systems. However, these processes are complex, and no single universally accepted method exists for carrying them out. Instead, the literature contains numerous proposals aimed at optimizing system performance by refining the execution of these tasks. In this work, to provide tools for optimizing service and resource allocation, SimulateIoT, an IoT simulator based on Model-Driven Development, has been extended to support these concepts. This extension, named SimulateIoT-Services, enables users to model, validate, generate, deploy, and test IoT systems, assessing their performance and evaluating how different service and resource allocation strategies impact overall system performance. Finally, this study not only extends SimulateIoT towards SimulateIot-Services, but also showcases its practical application. Namely, in the design and evaluation of scalable IoT systems and services through a case study centered on the IoV and efficient parking within a smart city context.</div></div>","PeriodicalId":29968,"journal":{"name":"Internet of Things","volume":"34 ","pages":"Article 101736"},"PeriodicalIF":7.6,"publicationDate":"2025-08-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144921290","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":"Fusion-based congestion control method for the power internet of things combining data-driven and rule-engine models","authors":"Chunxin Wang ,&nbsp;Wenyu Qu ,&nbsp;Rui Hou ,&nbsp;Feng Jiao ,&nbsp;Ying Zou","doi":"10.1016/j.iot.2025.101738","DOIUrl":"10.1016/j.iot.2025.101738","url":null,"abstract":"<div><div>The integration of the Internet of Things (IoT) into smart grids is placing an ever-growing emphasis on achieving high transmission stability. With the massive integration of heterogeneous devices into the power IoT, the volume of transmitted data has surged, and the dynamic complexity of networks has increased, leading to prolonged data congestion due to delayed transmission. Existing congestion control algorithms are typically divided into rule-based and learning-based approaches. However, relying exclusively on a single type of algorithm can result in challenges like suboptimal bandwidth utilization or slow convergence. To enhance data transmission stability, this paper introduces a congestion control method that integrates rule-based models with data-driven models. The method integrates network temporal features extracted by the Transformer algorithm with rule-based features from Google's Congestion Control (GCC) algorithm, and employs Proximal Policy Optimization (PPO) reinforcement learning to generate real-time transmission rate control strategies. This approach improves both the efficiency and robustness of deep reinforcement learning algorithms for congestion control. The performance evaluation metrics used for comparison are Quality of Service (QoS), mainly focusing on throughput, latency, and average packet loss rate. Compared to other algorithms, this model achieves the best performance with a score of 88.13. The hybrid algorithm offers a promising direction for future network congestion control.</div></div>","PeriodicalId":29968,"journal":{"name":"Internet of Things","volume":"34 ","pages":"Article 101738"},"PeriodicalIF":7.6,"publicationDate":"2025-08-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144912175","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":"Efficient IoT indoor monitoring via distributed deep learning in hybrid VLC/RF network architectures","authors":"Thai-Ha Dang ,&nbsp;Ngoc-Hai Dang ,&nbsp;Viet-Thang Tran","doi":"10.1016/j.iot.2025.101715","DOIUrl":"10.1016/j.iot.2025.101715","url":null,"abstract":"<div><div>The rapid growth of the Internet of Things (IoT) has driven the demand for efficient indoor monitoring systems that can provide reliable connectivity, low latency, and high energy efficiency. Traditional wireless approaches based solely on radio frequency (RF) suffer from spectrum congestion and interference, while visible light communication (VLC) systems are constrained by line-of-sight requirements and limited coverage. To address these challenges, this paper proposes a hybrid VLC/RF network architecture integrated with distributed deep learning for IoT-based indoor monitoring. In the proposed framework, IoT devices leverage VLC links for high-speed data transmission in favorable conditions and seamlessly switch to RF links to maintain connectivity in non-line-of-sight scenarios. A distributed deep learning framework is deployed across edge nodes to enable scalable and privacy-preserving analytics, reducing reliance on centralized processing while improving adaptability to dynamic indoor environments. Experimental evaluation demonstrates that the proposed system achieves higher monitoring accuracy, reduced latency, and improved energy efficiency compared to single-mode VLC or RF systems. These findings highlight the potential of hybrid communication networks combined with distributed intelligence to enhance the performance and robustness of IoT indoor monitoring applications in smart homes, healthcare, and industrial environments.</div></div>","PeriodicalId":29968,"journal":{"name":"Internet of Things","volume":"34 ","pages":"Article 101715"},"PeriodicalIF":7.6,"publicationDate":"2025-08-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144912176","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":"How intelligence is reshaping today: IoE edge networks?","authors":"Syed Usman Jamil ,&nbsp;M. Arif Khan ,&nbsp;M.A. Rahman ,&nbsp;Muhammad Ali Paracha ,&nbsp;Tanveer Zia ,&nbsp;Syed Sadiqur Rahman ,&nbsp;Syed Bilal Ahmed","doi":"10.1016/j.iot.2025.101717","DOIUrl":"10.1016/j.iot.2025.101717","url":null,"abstract":"<div><div>The rapid advancement of Internet of Everything (IoE) devices, coupled with emergent communication technologies such as Sixth Generation (6G), is facilitating the development of IoE-based edge networks wherein services are rendered closer to the network periphery. In these networks, devices exchange information and share computational resources. This research paper presents a novel methodological framework designed to optimise task allocation and device selection in IoE-6G systems. By integrating intelligent algorithms, our study investigates the impact of device selection delay, computational efficiency, and task success rates on overall system performance. The unique capabilities of 6G, such as AI-native infrastructure, Ultra-Reliable Low-Latency Communication (URLLC), Massive Machine-Type Communication (mMTC), Terahertz (THz) frequency bands, and dynamic network slicing, form the foundational enablers of our proposed framework. These features are essential for scalable and real-time IoE operations and are tightly integrated into our proposed system’s algorithmic design. Our methodological framework involves extensive simulations that evaluate the proposed system across various scenarios, focusing on foundational concepts and performance metrics that are essential for understanding the parameters influencing our research outcomes. We provide a detailed comparison of traditional and intelligent scheduling algorithms, showcasing significant improvements in task allocation and completion times when intelligence is employed. The novelty of Intelligent Main Task Off-loading Scheduling Algorithm (<em>i</em>MTOSA) lies in its dynamic intelligence and adaptive scheduling, optimising IoE task offloading with minimal communication and enhanced scalability by focusing on advanced groups of metrics and thoroughly discussing the implications for real-world IoE-6G environments. Our results contribute to a deeper understanding of the integration of intelligent systems in modern communication networks, paving the way for future advancements in IoE technologies. In overall performance, the intelligent variants of the proposed algorithm show less than 50% affected tasks, while non-intelligent scheduling algorithms exceed 90% affected tasks.</div></div>","PeriodicalId":29968,"journal":{"name":"Internet of Things","volume":"33 ","pages":"Article 101717"},"PeriodicalIF":7.6,"publicationDate":"2025-08-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144865692","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":"An explainable three-way neural network approach for intrusion detection in IoT ecosystem","authors":"Fazal Wahab ,&nbsp;Shengjun Ma ,&nbsp;Yuhai Zhao ,&nbsp;Anwar Shah","doi":"10.1016/j.iot.2025.101722","DOIUrl":"10.1016/j.iot.2025.101722","url":null,"abstract":"<div><div>Deep neural networks (DNNs) have demonstrated remarkable potential in intrusion detection; however, the inherent uncertainty in the data often results in false alerts due to the deterministic nature of their probabilistic classification mechanism. In IoT environments, this can affect system reliability. Moreover, most of the DNN-based methods do not provide a transparent and interpretable output of the model. A secure, reliable, and accurate framework is urgently required to protect IoT systems. To fill this gap, this article introduces a novel, explainable three-way neural network approach called Ex3WNN. We introduce a 3WC mechanism in the final layer of the DNN using Game-theoretic Rough Sets (GTRS). This allows the model to handle uncertain cases more intelligently by distinguishing between confident decisions and those requiring further interpretation. By categorizing predictions into attack, suspicious (uncertain), and normal classes, 3WC helps manage uncertainty, ensuring that ambiguous cases are flagged for further analysis rather than misclassified. The GTRS framework is employed to determine optimal decision thresholds, which are derived while achieving the trade-off between generality and accuracy. This approach enhances both detection accuracy and reliability. Incorporating the suspicious region can considerably reduce false alerts and significantly enhance the reliability, security, and confidence of the intrusion detection system (IDS). Furthermore, we utilize the eXplainable AI (XAI) techniques to provide an interpretable and transparent model’s output. The experimental results from four relevant and comprehensive datasets show that the proposed method outperformed existing baselines.</div></div>","PeriodicalId":29968,"journal":{"name":"Internet of Things","volume":"33 ","pages":"Article 101722"},"PeriodicalIF":7.6,"publicationDate":"2025-08-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144851979","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":"The evolving threat landscape of botnets: Comprehensive analysis of detection techniques in the age of artificial intelligence","authors":"Arash Mahboubi ,&nbsp;Khanh Luong ,&nbsp;Hamed Aboutorab ,&nbsp;Hang Thanh Bui ,&nbsp;Seyit Camtepe ,&nbsp;Keyvan Ansari ,&nbsp;Bazara Barry","doi":"10.1016/j.iot.2025.101728","DOIUrl":"10.1016/j.iot.2025.101728","url":null,"abstract":"<div><div>Botnets represent a significant and evolving cybersecurity threat, leveraging networks of compromised devices for various malicious activities, including data exfiltration (e.g., Truebot malware), credential theft, and distributed denial-of-service (DDoS) attacks. heir increasing sophistication includes advanced evasion techniques such as domain generation algorithms (DGAs), encrypted command-and-control (C&amp;C) channels, and peer-to-peer (P2P) architectures. These innovations pose substantial challenges to conventional detection systems. Existing surveys typically examine isolated detection methodologies or specific datasets, failing to address comprehensively the broader landscape, especially regarding adversarial manipulation of machine learning (ML) and artificial intelligence (AI) feature sets. To address this critical gap, this survey introduces the first systematic adversarial-aware analysis of botnet detection strategies. It specifically evaluates how adversaries exploit ML/AI feature manipulation, such as through noise injection and feature perturbation, to evade detection, a perspective that has not been quantitatively addressed in prior literature. A core contribution is our explicit benchmarking of detection model robustness across four quantitative metrics, faithfulness, monotonicity, sensitivity, and complexity, providing novel insights into the resilience of state-of-the-art models under adversarial conditions. Additionally, we highlight persistent practical challenges including limited dataset diversity and dependence on high-quality labeled data, and propose potential mitigation approaches such as synthetic data generation, federated and semi-supervised learning, and lightweight detection architectures tailored for resource-constrained IoT deployments. Finally, we outline key future research directions emphasizing standardized robustness evaluation frameworks, explainable AI to enhance interpretability and trust, and privacy-preserving collaborative data-sharing mechanisms. By integrating this adversarial-aware perspective with a comprehensive and practical evaluation framework, this work contributes to the field’s understanding of botnet detection and supports the design of more robust and resilient cybersecurity solutions through insights relevant to both researchers and practitioners.</div></div>","PeriodicalId":29968,"journal":{"name":"Internet of Things","volume":"33 ","pages":"Article 101728"},"PeriodicalIF":7.6,"publicationDate":"2025-08-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144827133","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":"Business continuity of Cloud-based IoT applications through a seamless continuum","authors":"Carmine Colarusso ,&nbsp;Ida Falco ,&nbsp;Eugenio Zimeo","doi":"10.1016/j.iot.2025.101723","DOIUrl":"10.1016/j.iot.2025.101723","url":null,"abstract":"<div><div>IoT systems often exploit the Cloud to process and share collected data, while pre-processing at the Edge reduces the amount of data transferred to the network and processed in the Cloud. Despite the relevant advantages of using centralized and multiplexed high-performance resources, Edge-Cloud interactions might not be possible due to transient network failures, so negatively impacting business continuity. In this paper, we extend the adoption of Edge computing by seamlessly mirroring in Edge, through a framework, some Cloud-deployed components of the business logic and related data of IoT applications to improve service availability when transient network failures impede Edge-Cloud communication. In particular, to address potential inconsistency due to data replication among Edge nodes and Cloud, we propose a new distributed consistency algorithm, called EdgeCloudWPaxos. It is based on a distributed, event-driven technique that implements reconciliation when distributed local transactions occur. The paper presents the framework, its adoption for wrapping an existing application for item tracking, and the validation through several tests conducted by emulating the tracking activity produced by a number of retail shops, which shows the effectiveness of the framework in ensuring business continuity and lower response times even in the case of network failure.</div></div>","PeriodicalId":29968,"journal":{"name":"Internet of Things","volume":"33 ","pages":"Article 101723"},"PeriodicalIF":7.6,"publicationDate":"2025-08-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144865690","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":"IoT-based monitoring and control for optimized plant growth in smart greenhouses using soil and hydroponic systems","authors":"Kenza Bouarroudj ,&nbsp;Fatima Babaa ,&nbsp;Abderrahim Touil","doi":"10.1016/j.iot.2025.101710","DOIUrl":"10.1016/j.iot.2025.101710","url":null,"abstract":"&lt;div&gt;&lt;h3&gt;Context:&lt;/h3&gt;&lt;div&gt;Agriculture is under mounting pressure from climate change, natural resource depletion, and the urgent need for global food security. Smart technologies, particularly IoT-enabled greenhouse systems, offer a promising pathway to sustainably intensify crop production. However, achieving seamless real-time monitoring, autonomous control, and fault detection remains technically complex, especially in remote or off-grid regions with limited infrastructure and unstable energy supply.&lt;/div&gt;&lt;/div&gt;&lt;div&gt;&lt;h3&gt;Objective:&lt;/h3&gt;&lt;div&gt;This study aims to develop a fully autonomous, intelligent greenhouse system that integrates real-time environmental monitoring, adaptive control, and embedded fault diagnosis. A key focus is on enabling continuous and reliable operation in off-grid conditions through solar energy autonomy. The system is designed to enhance crop productivity, energy efficiency, and resilience across a range of agricultural contexts, from smallholder plots to commercial-scale operations.&lt;/div&gt;&lt;/div&gt;&lt;div&gt;&lt;h3&gt;Methods:&lt;/h3&gt;&lt;div&gt;The system integrates a distributed sensor network with a centralized control platform and mobile interface to monitor key agronomic and technical variables, including temperature, humidity, &lt;span&gt;&lt;math&gt;&lt;mrow&gt;&lt;mi&gt;C&lt;/mi&gt;&lt;msub&gt;&lt;mrow&gt;&lt;mi&gt;O&lt;/mi&gt;&lt;/mrow&gt;&lt;mrow&gt;&lt;mn&gt;2&lt;/mn&gt;&lt;/mrow&gt;&lt;/msub&gt;&lt;/mrow&gt;&lt;/math&gt;&lt;/span&gt;concentration, light intensity, irrigation flow, nutrient composition, and electrical parameters. To enable predictive maintenance, an anomaly detection module based on the Isolation Forest algorithm was implemented and trained on synthetically generated multivariate time series data representing realistic fault scenarios (e.g., irradiance drop, overheating, sensor drift, voltage imbalance). The algorithm’s performance was quantitatively assessed using confusion matrices, receiver operating characteristic (ROC) curves, and classification metrics, achieving high anomaly detection accuracy. Predefined alert thresholds were assigned to each monitored parameter and integrated with machine learning outputs to enhance diagnostic reliability. The system also features a solar energy harvesting subsystem with continuous tracking of photovoltaic voltage, current, energy yield, and battery state-of-charge, supporting fully autonomous, off-grid operation.&lt;/div&gt;&lt;/div&gt;&lt;div&gt;&lt;h3&gt;Results and Conclusions:&lt;/h3&gt;&lt;div&gt;Controlled-environment testing demonstrated the system’s ability to autonomously and precisely regulate greenhouse climate parameters while ensuring operational continuity under simulated fault and energy fluctuation scenarios. The integrated fault detection module showed high diagnostic accuracy, and visual analytics were incorporated to support interpretability by non-specialist users. Although still in the pre-deployment phase, these results confirm the system’s technical robustness and readiness for field trials. A comparative analysis with existing solutions emphasizes the system’s unique contribution—namely, the ","PeriodicalId":29968,"journal":{"name":"Internet of Things","volume":"33 ","pages":"Article 101710"},"PeriodicalIF":7.6,"publicationDate":"2025-08-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144860358","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":"An efficient and secure adaptive federated learning method based on CKKS for data processing in the Internet of Things","authors":"Yang Lan ,&nbsp;Lixiang Li ,&nbsp;Haipeng Peng ,&nbsp;Yeqing Ren ,&nbsp;Zhongkai Dang","doi":"10.1016/j.iot.2025.101725","DOIUrl":"10.1016/j.iot.2025.101725","url":null,"abstract":"<div><div>Federated learning (FL) provides a new paradigm for solving the security of private data in the internet of things (IoT). However, the huge consumption of computing resources and communication cost makes the FL process inefficient. To solve above problems, this paper proposes an efficient and secure adaptive federated learning method based on CKKS homomorphic encryption (HE) for data processing in IoT. Inspired by dropout, we propose an adaptive inactivation of weights strategy. Through adaptive change of inactivation parameter, part of the weights after reorganization are encrypted and uploaded in each communication. The dual protection of reorganization operation and HE can better protect the weights information. Then, to alleviate the impact of the above methods on the performance of FL, the local data distribution and the change of model accuracy are considered, we propose the federated aggregation method with reward and punishment factor, and the historical information of the local model is employed to design a weights correction strategy. Finally, we use MNIST dataset, fashion-MNIST dataset, GTSRB dataset and CSE-CIC-IDS2018 dataset to design non-independent and identically distributed data scenarios, and a large number of experiments are carried out to verify the effectiveness of the proposed method. Our method not only protects the privacy of weights information, but also reduces the communication cost and the local resource consumption caused by the encryption, which provides a good reference for the follow-up development of FL in IoT.</div></div>","PeriodicalId":29968,"journal":{"name":"Internet of Things","volume":"33 ","pages":"Article 101725"},"PeriodicalIF":7.6,"publicationDate":"2025-08-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144851980","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}