The convergence of Artificial Intelligence(AI)and the Internet of Things(IoT)has enabled Artificial Intelligence of Things(AIoT)systems that support intelligent and responsive smart societies,but it also introduces ma...The convergence of Artificial Intelligence(AI)and the Internet of Things(IoT)has enabled Artificial Intelligence of Things(AIoT)systems that support intelligent and responsive smart societies,but it also introduces major security and privacy concerns across domains such as healthcare,transportation,and smart cities.This Systemic Literature Review(SLR)addresses three research questions:identifying major threats and challenges in AIoT ecosystems,reviewing state-of-the-art security and privacy techniques,and evaluating their effectiveness.An SLR covering the period from 2020 to 2025 was conducted using major academic digital libraries,including IEEE Xplore,ACM Digital Library,ScienceDirect,SpringerLink,and Wiley Online Library,with a focus on security-and privacy-enhancing techniques such as blockchain,federated learning,and edge AI.The SLR identifies key challenges including data privacy leakage,authentication,cloud dependency,and attack surface expansion,and finds that emerging techniques,while promising,often involve trade-offs related to latency,scalability,and compliance.The study highlights future directions including lightweight cryptography,standardization,and explainable AI to support secure and trustworthy AIoT-enabled smart societies.展开更多
With the rapid development of the Artificial Intelligence of Things(AIoT),convolutional neural networks(CNNs)have demonstrated potential and remarkable performance in AIoT applications due to their excellent performan...With the rapid development of the Artificial Intelligence of Things(AIoT),convolutional neural networks(CNNs)have demonstrated potential and remarkable performance in AIoT applications due to their excellent performance in various inference tasks.However,the users have concerns about privacy leakage for the use of AI and the performance and efficiency of computing on resource-constrained IoT edge devices.Therefore,this paper proposes an efficient privacy-preserving CNN framework(i.e.,EPPA)based on the Fully Homomorphic Encryption(FHE)scheme for AIoT application scenarios.In the plaintext domain,we verify schemes with different activation structures to determine the actual activation functions applicable to the corresponding ciphertext domain.Within the encryption domain,we integrate batch normalization(BN)into the convolutional layers to simplify the computation process.For nonlinear activation functions,we use composite polynomials for approximate calculation.Regarding the noise accumulation caused by homomorphic multiplication operations,we realize the refreshment of ciphertext noise through minimal“decryption-encryption”interactions,instead of adopting bootstrapping operations.Additionally,in practical implementation,we convert three-dimensional convolution into two-dimensional convolution to reduce the amount of computation in the encryption domain.Finally,we conduct extensive experiments on four IoT datasets,different CNN architectures,and two platforms with different resource configurations to evaluate the performance of EPPA in detail.展开更多
针对沙滩水域环境复杂且难以有效清洁的问题,综合应用人工智能物联网(Artificial Internet of Things, AIoT)和视觉识别等先进技术,开发了一款具有远程控制、视觉识别、智能抓取、重量检测及智能显示等功能的多功能清洁机器人。该机器...针对沙滩水域环境复杂且难以有效清洁的问题,综合应用人工智能物联网(Artificial Internet of Things, AIoT)和视觉识别等先进技术,开发了一款具有远程控制、视觉识别、智能抓取、重量检测及智能显示等功能的多功能清洁机器人。该机器人专为提升清洁效率和自动化水平设计,配备了远程控制、视觉识别、智能抓取、重量检测及状态显示等功能。采用英伟达Jetson Nano作为核心处理器,结合Intel D415深度相机和基于FloW数据集训练的YOLOv8算法,实现水面漂浮垃圾的实时检测与精确定位。系统通过STM32微控制器解析视觉数据并控制机械臂完成精准抓取。为提高移动性能,机器人采用麦克纳姆轮实现全向运动,当内置称重传感器检测到收集装置满载时,系统可自主返回基地卸载垃圾。此外,系统集成HC-05蓝牙模块实现远程无线控制,并通过OLED显示屏实时显示工作状态。综合应用了AIoT、自动化控制及视觉识别技术,突破了传统清洁方式的局限,显著提升了沙滩水域清洁工作的效率和便捷性,为环保行动提供了强有力的工具。展开更多
针对云南省通海县蔬菜供应链中存在的产销信息孤岛效应显著、非结构化数据处理滞后以及冷链物流断链导致的损耗率居高不下等问题,提出一种基于人工智能物联网(Artificial Intelligence of Things,AIoT)技术的农业适配型供应链协同管理...针对云南省通海县蔬菜供应链中存在的产销信息孤岛效应显著、非结构化数据处理滞后以及冷链物流断链导致的损耗率居高不下等问题,提出一种基于人工智能物联网(Artificial Intelligence of Things,AIoT)技术的农业适配型供应链协同管理系统架构。该系统构建“端—边—云”三级协同架构,通过边缘侧数据清洗抑制高原强干扰环境下的传感器噪声,利用改进Arrhenius方程建立适配通海叶菜高呼吸速率特性的货架期预测模型,并以联盟链实现多主体可信溯源。在通海至广州典型线路的3个月试运行中,叶菜平均腐损率由12.4%降至4.6%,订单平均响应时间缩短约63.1%,单车平均净利润提高约39.3%。归因分析表明,损耗降低主要归功于边缘预警与动态调度的协同作用。该系统为高原特色农业数字化转型提供了兼顾技术适配与社会嵌入的工程化方案。展开更多
随着信息技术的飞速发展,人工智能与物联网(Artificial Intelligence of Things,AIoT)技术逐渐成为工程造价管理领域的新兴力量。文章深入探讨了AIoT技术在施工过程造价实时监控中的应用,旨在通过智能化的数据采集、处理与分析,提升造...随着信息技术的飞速发展,人工智能与物联网(Artificial Intelligence of Things,AIoT)技术逐渐成为工程造价管理领域的新兴力量。文章深入探讨了AIoT技术在施工过程造价实时监控中的应用,旨在通过智能化的数据采集、处理与分析,提升造价监控的效率与准确性,推动工程造价管理的智能化转型,为工程造价管理提供了新的思路和方法。展开更多
Recently,the Internet of Things(IoT)technology has been utilized in a wide range of services and applications which significantly transforms digital ecosystems through seamless interconnectivity between various smart ...Recently,the Internet of Things(IoT)technology has been utilized in a wide range of services and applications which significantly transforms digital ecosystems through seamless interconnectivity between various smart devices.Furthermore,the IoT plays a key role in multiple domains,including industrial automation,smart homes,and intelligent transportation systems.However,an increasing number of connected devices presents significant challenges related to efficient resource allocation and system responsiveness.To address these issue,this research proposes a Modified Walrus Optimization Algorithm(MWaOA)for effective resource management in smart IoT systems.In the proposed MWaOA,a crowding process is incorporated to maintain diversity and avoid premature convergence thereby enhancing the global search capability.During resource allocation,the MWaOA prevents early convergence,which aids in achieving a better balance between the exploration and exploitation phases during optimization.Empirical evaluations show that the MWaOA reduces energy consumption by approximately 4% to 34%and minimizes the response time by 6% to 33% across different service arrival rates.Compared to traditional optimization algorithms,MWaOA reduces energy consumption by 5% to 30%and minimizes the response time by 4% to 28% across different simulation epochs.The proposed MWaOA provides adaptive and robust resource allocation,thereby minimizing transmission cost while considering network constraints and real-time performance parameters.展开更多
Deploying Large LanguageModel(LLM)-based agents in the Industrial Internet ofThings(IIoT)presents significant challenges,including high latency from cloud-based APIs,data privacy concerns,and the infeasibility of depl...Deploying Large LanguageModel(LLM)-based agents in the Industrial Internet ofThings(IIoT)presents significant challenges,including high latency from cloud-based APIs,data privacy concerns,and the infeasibility of deploying monolithic models on resource-constrained edge devices.While smaller models(SLMs)are suitable for edge deployment,they often lack the reasoning power for complex,multi-step tasks.To address these issues,this paper introduces LEAF,a Lightweight Edge Agent Framework designed for efficiently executing complex tasks at the edge.LEAF employs a novel architecture where multiple expert SLMs—specialized for planning,execution,and interaction—work in concert,decomposing complex problems into manageable sub-tasks.To mitigate the resource overhead of this multi-model approach,LEAF implements an efficient parameter-sharing scheme based on Scalable Low-Rank Adaptation(S-LoRA).We introduce a two-stage training strategy combining Supervised Fine-Tuning(SFT)and Group Relative Policy Optimization(GRPO)to significantly enhance each expert’s capabilities.Furthermore,a Finite StateMachine(FSM)-based decision engine orchestrates the workflow,uniquely balancing deterministic control with intelligent flexibility,making it ideal for industrial environments that demand both reliability and adaptability.Experiments across diverse IIoT scenarios demonstrate that LEAF significantly outperforms baseline methods in both task success rate and user satisfaction.Notably,our fine-tuned 4-billion-parameter model achieves a task success rate over 90%in complex IIoT scenarios,demonstrating LEAF’s ability to deliver powerful and efficient autonomy at the industrial edge.展开更多
With more and more IoT terminals being deployed in various power grid business scenarios,terminal reliability has become a practical challenge that threatens the current security protection architecture.Most IoT termi...With more and more IoT terminals being deployed in various power grid business scenarios,terminal reliability has become a practical challenge that threatens the current security protection architecture.Most IoT terminals have security risks and vulnerabilities,and limited resources make it impossible to deploy costly security protection methods on the terminal.In order to cope with these problems,this paper proposes a lightweight trust evaluation model TCL,which combines three network models,TCN,CNN,and LSTM,with stronger feature extraction capability and can score the reliability of the device by periodically analyzing the traffic behavior and activity logs generated by the terminal device,and the trust evaluation of the terminal’s continuous behavior can be achieved by combining the scores of different periods.After experiments,it is proved that TCL can effectively use the traffic behaviors and activity logs of terminal devices for trust evaluation and achieves F1-score of 95.763,94.456,99.923,and 99.195 on HDFS,BGL,N-BaIoT,and KDD99 datasets,respectively,and the size of TCL is only 91KB,which can achieve similar or better performance than CNN-LSTM,RobustLog and other methods with less computational resources and storage space.展开更多
TheIndustrial Internet of Things(IIoT)has emerged as a cornerstone of Industry 4.0,enabling large-scale automation and data-driven decision-making across factories,supply chains,and critical infrastructures.However,th...TheIndustrial Internet of Things(IIoT)has emerged as a cornerstone of Industry 4.0,enabling large-scale automation and data-driven decision-making across factories,supply chains,and critical infrastructures.However,the massive interconnection of resource-constrained devices also amplifies the risks of eavesdropping,data tampering,and device impersonation.While digital signatures are indispensable for ensuring authenticity and non-repudiation,conventional schemes such as RSA and ECCare vulnerable to quantumalgorithms,jeopardizing long-termtrust in IIoT deployments.This study proposes a lightweight,stateless,hash-based signature scheme that achieves post-quantum security while addressing the stringent efficiency demands of IIoT.The design introduces two key optimizations:(1)Forest ofRandomSubsets(FORS)onDemand,where subset secret keys are generated dynamically via a PseudoRandom Function(PRF),thereby minimizing storage overhead and eliminating key-reuse risks;and(2)Winternitz One-Time Signature Plus(WOTS+)partial hash-chain caching,which precomputes intermediate hash values at edge gateways,reducing device-side computations,latency,and energy consumption.The architecture integrates a multi-layerMerkle authentication tree(Merkle tree)and role-based delegation across sensors,gateways,and a Signature Authority Center(SAC),supporting scalable cross-site deployment and key rotation.Froma theoretical perspective,we establish a formal(Existential Unforgeability under Chosen Message Attack)EUF-CMA security proof using a game-based reduction framework.The proof demonstrates that any successful forgerymust reduce to breaking the underlying assumptions of PRF indistinguishability,(second)preimage resistance,or collision resistance,thus quantifying adversarial advantage and ensuring unforgeability.On the implementation side,our design achieves a balanced trade-off between postquantum security and lightweight performance,offering concrete deployment guidelines for real-time industrial systems.In summary,the proposed method contributes both practical system design and formal security guarantees,providing IIoT with a deployable signature substrate that enhances resilience against quantum-era threats and supports future extensions such as device attestation,group signatures,and anomaly detection.展开更多
The rapid expansion of the Internet of Things(IoT)and Edge Artificial Intelligence(AI)has redefined automation and connectivity acrossmodern networks.However,the heterogeneity and limited resources of IoT devices expo...The rapid expansion of the Internet of Things(IoT)and Edge Artificial Intelligence(AI)has redefined automation and connectivity acrossmodern networks.However,the heterogeneity and limited resources of IoT devices expose them to increasingly sophisticated and persistentmalware attacks.These adaptive and stealthy threats can evade conventional detection,establish remote control,propagate across devices,exfiltrate sensitive data,and compromise network integrity.This study presents a Software-Defined Internet of Things(SD-IoT)control-plane-based,AI-driven framework that integrates Gated Recurrent Units(GRU)and Long Short-TermMemory(LSTM)networks for efficient detection of evolving multi-vector,malware-driven botnet attacks.The proposed CUDA-enabled hybrid deep learning(DL)framework performs centralized real-time detection without adding computational overhead to IoT nodes.A feature selection strategy combining variable clustering,attribute evaluation,one-R attribute evaluation,correlation analysis,and principal component analysis(PCA)enhances detection accuracy and reduces complexity.The framework is rigorously evaluated using the N_BaIoT dataset under k-fold cross-validation.Experimental results achieve 99.96%detection accuracy,a false positive rate(FPR)of 0.0035%,and a detection latency of 0.18 ms,confirming its high efficiency and scalability.The findings demonstrate the framework’s potential as a robust and intelligent security solution for next-generation IoT ecosystems.展开更多
文摘The convergence of Artificial Intelligence(AI)and the Internet of Things(IoT)has enabled Artificial Intelligence of Things(AIoT)systems that support intelligent and responsive smart societies,but it also introduces major security and privacy concerns across domains such as healthcare,transportation,and smart cities.This Systemic Literature Review(SLR)addresses three research questions:identifying major threats and challenges in AIoT ecosystems,reviewing state-of-the-art security and privacy techniques,and evaluating their effectiveness.An SLR covering the period from 2020 to 2025 was conducted using major academic digital libraries,including IEEE Xplore,ACM Digital Library,ScienceDirect,SpringerLink,and Wiley Online Library,with a focus on security-and privacy-enhancing techniques such as blockchain,federated learning,and edge AI.The SLR identifies key challenges including data privacy leakage,authentication,cloud dependency,and attack surface expansion,and finds that emerging techniques,while promising,often involve trade-offs related to latency,scalability,and compliance.The study highlights future directions including lightweight cryptography,standardization,and explainable AI to support secure and trustworthy AIoT-enabled smart societies.
基金supported by the Natural Science Foundation of China No.62362008the Major Scientific and Technological Special Project of Guizhou Province([2024]014).
文摘With the rapid development of the Artificial Intelligence of Things(AIoT),convolutional neural networks(CNNs)have demonstrated potential and remarkable performance in AIoT applications due to their excellent performance in various inference tasks.However,the users have concerns about privacy leakage for the use of AI and the performance and efficiency of computing on resource-constrained IoT edge devices.Therefore,this paper proposes an efficient privacy-preserving CNN framework(i.e.,EPPA)based on the Fully Homomorphic Encryption(FHE)scheme for AIoT application scenarios.In the plaintext domain,we verify schemes with different activation structures to determine the actual activation functions applicable to the corresponding ciphertext domain.Within the encryption domain,we integrate batch normalization(BN)into the convolutional layers to simplify the computation process.For nonlinear activation functions,we use composite polynomials for approximate calculation.Regarding the noise accumulation caused by homomorphic multiplication operations,we realize the refreshment of ciphertext noise through minimal“decryption-encryption”interactions,instead of adopting bootstrapping operations.Additionally,in practical implementation,we convert three-dimensional convolution into two-dimensional convolution to reduce the amount of computation in the encryption domain.Finally,we conduct extensive experiments on four IoT datasets,different CNN architectures,and two platforms with different resource configurations to evaluate the performance of EPPA in detail.
文摘针对沙滩水域环境复杂且难以有效清洁的问题,综合应用人工智能物联网(Artificial Internet of Things, AIoT)和视觉识别等先进技术,开发了一款具有远程控制、视觉识别、智能抓取、重量检测及智能显示等功能的多功能清洁机器人。该机器人专为提升清洁效率和自动化水平设计,配备了远程控制、视觉识别、智能抓取、重量检测及状态显示等功能。采用英伟达Jetson Nano作为核心处理器,结合Intel D415深度相机和基于FloW数据集训练的YOLOv8算法,实现水面漂浮垃圾的实时检测与精确定位。系统通过STM32微控制器解析视觉数据并控制机械臂完成精准抓取。为提高移动性能,机器人采用麦克纳姆轮实现全向运动,当内置称重传感器检测到收集装置满载时,系统可自主返回基地卸载垃圾。此外,系统集成HC-05蓝牙模块实现远程无线控制,并通过OLED显示屏实时显示工作状态。综合应用了AIoT、自动化控制及视觉识别技术,突破了传统清洁方式的局限,显著提升了沙滩水域清洁工作的效率和便捷性,为环保行动提供了强有力的工具。
文摘针对云南省通海县蔬菜供应链中存在的产销信息孤岛效应显著、非结构化数据处理滞后以及冷链物流断链导致的损耗率居高不下等问题,提出一种基于人工智能物联网(Artificial Intelligence of Things,AIoT)技术的农业适配型供应链协同管理系统架构。该系统构建“端—边—云”三级协同架构,通过边缘侧数据清洗抑制高原强干扰环境下的传感器噪声,利用改进Arrhenius方程建立适配通海叶菜高呼吸速率特性的货架期预测模型,并以联盟链实现多主体可信溯源。在通海至广州典型线路的3个月试运行中,叶菜平均腐损率由12.4%降至4.6%,订单平均响应时间缩短约63.1%,单车平均净利润提高约39.3%。归因分析表明,损耗降低主要归功于边缘预警与动态调度的协同作用。该系统为高原特色农业数字化转型提供了兼顾技术适配与社会嵌入的工程化方案。
文摘随着信息技术的飞速发展,人工智能与物联网(Artificial Intelligence of Things,AIoT)技术逐渐成为工程造价管理领域的新兴力量。文章深入探讨了AIoT技术在施工过程造价实时监控中的应用,旨在通过智能化的数据采集、处理与分析,提升造价监控的效率与准确性,推动工程造价管理的智能化转型,为工程造价管理提供了新的思路和方法。
文摘Recently,the Internet of Things(IoT)technology has been utilized in a wide range of services and applications which significantly transforms digital ecosystems through seamless interconnectivity between various smart devices.Furthermore,the IoT plays a key role in multiple domains,including industrial automation,smart homes,and intelligent transportation systems.However,an increasing number of connected devices presents significant challenges related to efficient resource allocation and system responsiveness.To address these issue,this research proposes a Modified Walrus Optimization Algorithm(MWaOA)for effective resource management in smart IoT systems.In the proposed MWaOA,a crowding process is incorporated to maintain diversity and avoid premature convergence thereby enhancing the global search capability.During resource allocation,the MWaOA prevents early convergence,which aids in achieving a better balance between the exploration and exploitation phases during optimization.Empirical evaluations show that the MWaOA reduces energy consumption by approximately 4% to 34%and minimizes the response time by 6% to 33% across different service arrival rates.Compared to traditional optimization algorithms,MWaOA reduces energy consumption by 5% to 30%and minimizes the response time by 4% to 28% across different simulation epochs.The proposed MWaOA provides adaptive and robust resource allocation,thereby minimizing transmission cost while considering network constraints and real-time performance parameters.
文摘Deploying Large LanguageModel(LLM)-based agents in the Industrial Internet ofThings(IIoT)presents significant challenges,including high latency from cloud-based APIs,data privacy concerns,and the infeasibility of deploying monolithic models on resource-constrained edge devices.While smaller models(SLMs)are suitable for edge deployment,they often lack the reasoning power for complex,multi-step tasks.To address these issues,this paper introduces LEAF,a Lightweight Edge Agent Framework designed for efficiently executing complex tasks at the edge.LEAF employs a novel architecture where multiple expert SLMs—specialized for planning,execution,and interaction—work in concert,decomposing complex problems into manageable sub-tasks.To mitigate the resource overhead of this multi-model approach,LEAF implements an efficient parameter-sharing scheme based on Scalable Low-Rank Adaptation(S-LoRA).We introduce a two-stage training strategy combining Supervised Fine-Tuning(SFT)and Group Relative Policy Optimization(GRPO)to significantly enhance each expert’s capabilities.Furthermore,a Finite StateMachine(FSM)-based decision engine orchestrates the workflow,uniquely balancing deterministic control with intelligent flexibility,making it ideal for industrial environments that demand both reliability and adaptability.Experiments across diverse IIoT scenarios demonstrate that LEAF significantly outperforms baseline methods in both task success rate and user satisfaction.Notably,our fine-tuned 4-billion-parameter model achieves a task success rate over 90%in complex IIoT scenarios,demonstrating LEAF’s ability to deliver powerful and efficient autonomy at the industrial edge.
基金supported by National Key R&D Program of China(No.2022YFB3105101).
文摘With more and more IoT terminals being deployed in various power grid business scenarios,terminal reliability has become a practical challenge that threatens the current security protection architecture.Most IoT terminals have security risks and vulnerabilities,and limited resources make it impossible to deploy costly security protection methods on the terminal.In order to cope with these problems,this paper proposes a lightweight trust evaluation model TCL,which combines three network models,TCN,CNN,and LSTM,with stronger feature extraction capability and can score the reliability of the device by periodically analyzing the traffic behavior and activity logs generated by the terminal device,and the trust evaluation of the terminal’s continuous behavior can be achieved by combining the scores of different periods.After experiments,it is proved that TCL can effectively use the traffic behaviors and activity logs of terminal devices for trust evaluation and achieves F1-score of 95.763,94.456,99.923,and 99.195 on HDFS,BGL,N-BaIoT,and KDD99 datasets,respectively,and the size of TCL is only 91KB,which can achieve similar or better performance than CNN-LSTM,RobustLog and other methods with less computational resources and storage space.
文摘TheIndustrial Internet of Things(IIoT)has emerged as a cornerstone of Industry 4.0,enabling large-scale automation and data-driven decision-making across factories,supply chains,and critical infrastructures.However,the massive interconnection of resource-constrained devices also amplifies the risks of eavesdropping,data tampering,and device impersonation.While digital signatures are indispensable for ensuring authenticity and non-repudiation,conventional schemes such as RSA and ECCare vulnerable to quantumalgorithms,jeopardizing long-termtrust in IIoT deployments.This study proposes a lightweight,stateless,hash-based signature scheme that achieves post-quantum security while addressing the stringent efficiency demands of IIoT.The design introduces two key optimizations:(1)Forest ofRandomSubsets(FORS)onDemand,where subset secret keys are generated dynamically via a PseudoRandom Function(PRF),thereby minimizing storage overhead and eliminating key-reuse risks;and(2)Winternitz One-Time Signature Plus(WOTS+)partial hash-chain caching,which precomputes intermediate hash values at edge gateways,reducing device-side computations,latency,and energy consumption.The architecture integrates a multi-layerMerkle authentication tree(Merkle tree)and role-based delegation across sensors,gateways,and a Signature Authority Center(SAC),supporting scalable cross-site deployment and key rotation.Froma theoretical perspective,we establish a formal(Existential Unforgeability under Chosen Message Attack)EUF-CMA security proof using a game-based reduction framework.The proof demonstrates that any successful forgerymust reduce to breaking the underlying assumptions of PRF indistinguishability,(second)preimage resistance,or collision resistance,thus quantifying adversarial advantage and ensuring unforgeability.On the implementation side,our design achieves a balanced trade-off between postquantum security and lightweight performance,offering concrete deployment guidelines for real-time industrial systems.In summary,the proposed method contributes both practical system design and formal security guarantees,providing IIoT with a deployable signature substrate that enhances resilience against quantum-era threats and supports future extensions such as device attestation,group signatures,and anomaly detection.
基金supported by Princess Nourah bint Abdulrahman University Researchers Supporting ProjectNumber(PNURSP2025R97),PrincessNourah bint AbdulrahmanUniversity,Riyadh,Saudi Arabia.
文摘The rapid expansion of the Internet of Things(IoT)and Edge Artificial Intelligence(AI)has redefined automation and connectivity acrossmodern networks.However,the heterogeneity and limited resources of IoT devices expose them to increasingly sophisticated and persistentmalware attacks.These adaptive and stealthy threats can evade conventional detection,establish remote control,propagate across devices,exfiltrate sensitive data,and compromise network integrity.This study presents a Software-Defined Internet of Things(SD-IoT)control-plane-based,AI-driven framework that integrates Gated Recurrent Units(GRU)and Long Short-TermMemory(LSTM)networks for efficient detection of evolving multi-vector,malware-driven botnet attacks.The proposed CUDA-enabled hybrid deep learning(DL)framework performs centralized real-time detection without adding computational overhead to IoT nodes.A feature selection strategy combining variable clustering,attribute evaluation,one-R attribute evaluation,correlation analysis,and principal component analysis(PCA)enhances detection accuracy and reduces complexity.The framework is rigorously evaluated using the N_BaIoT dataset under k-fold cross-validation.Experimental results achieve 99.96%detection accuracy,a false positive rate(FPR)of 0.0035%,and a detection latency of 0.18 ms,confirming its high efficiency and scalability.The findings demonstrate the framework’s potential as a robust and intelligent security solution for next-generation IoT ecosystems.
