IEEE Transactions on Consumer Electronics最新文献

{"title":"CGCN-DF: A Cascade GCN Vulnerability Detection Framework for Secure CIoT Smart Contracts","authors":"Keqing Wang;Wenyin Zhang;Mengru Tu;Weihao Su","doi":"10.1109/TCE.2026.3664215","DOIUrl":"https://doi.org/10.1109/TCE.2026.3664215","url":null,"abstract":"With the widespread use of smart contracts in blockchain applications, particularly in the consumer Internet of Things (CIoT), the security of smart contracts has become increasingly critical. CIoT refers to an IoT environment where various consumer devices are intelligently connected and interact via the internet. In this context, smart contracts are frequently used to automate tasks such as device control, data sharing, and transaction verification. However, vulnerabilities in smart contracts can lead to system attacks, thereby compromising the security of the entire IoT network. The Cascade Graph Convolutional Network-based Vulnerability Detection Framework for smart contracts (CGCN-DF) proposed in this study effectively enhances the security assessment of smart contracts in CIoT applications. Devices and systems in CIoT environments typically exhibit high heterogeneity and complex interaction patterns, but a single graph structure cannot fully capture the multidimensional behavioral features and complex structural relationships of smart contracts. To address this issue, this paper introduces the Semantic Contract Graph (SCG), which integrates three graph representations—Abstract Syntax Tree (AST), Control Flow Graph (CFG), and Data Flow Graph (DFG)—into a unified graph structure, comprehensively covering different aspects of the code. Furthermore, as smart contracts in CIoT environments often involve real-time data flows and complex execution paths, the CGCN-DF framework employs a cascading mechanism that performs three-level graph convolutional processing through relational, Meta-path structures, and cyclic structures. This approach extracts rich and complementary information from multi-level features, explicitly models the dynamic interactions among syntax, execution paths, and data flows, and enhances the model’s contextual awareness of vulnerability-triggering conditions. Ultimately, the framework achieves coarse-grained detection at the contract level and fine-grained detection at the line level. Experimental results demonstrate that the proposed method can effectively localize vulnerabilities down to specific code lines. This not only enhances the precision and practicality of smart contract vulnerability detection but, more importantly, contributes an innovative technical framework and theoretical methodology to the CIoT field—particularly in areas such as DSL program analysis and the intersection of IoT and blockchain security. This work thereby helps advance the field toward more refined and context-aware security analysis aligned with the realistic characteristics of complex systems.","PeriodicalId":13208,"journal":{"name":"IEEE Transactions on Consumer Electronics","volume":"72 3","pages":"7148-7159"},"PeriodicalIF":9.9,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148836740","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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":"Intelligent Post-Quantum Cryptography for Secure Consumer Internet of Things Networks","authors":"Ishtiaq Ahmad;Liang Yang;Yazeed Alkhrijah;Mohamad A. Alawad;Ahmad Almadhor;Limei Peng","doi":"10.1109/TCE.2026.3671160","DOIUrl":"https://doi.org/10.1109/TCE.2026.3671160","url":null,"abstract":"The rapid growth of consumer Internet of Things (IoT) networks is transforming industries while introducing complex security challenges, particularly in the era of quantum computing, which threatens traditional encryption methods. Post-quantum cryptography (PQC) provides quantum-resistant protection, yet static deployment often fails to meet the dynamic and resource-constrained requirements of IoT systems. To address this, we propose an intelligent and adaptive security framework that integrates lattice-based PQC with a graph neural network (GNN) and a deterministic soft actor-critic (DSAC) algorithm. The GNN continuously monitors network activity to detect anomalies and provides real-time threat intelligence to the DSAC agent, which dynamically adjusts PQC parameters to optimize security. Simulation results demonstrate that the proposed framework can improve security by up to 40% while reducing computational cost and latency by 27% and 46%, respectively, highlighting the potential of Artificial Intelligence (AI)-driven, post-quantum cryptography for securing dynamic IoT networks.","PeriodicalId":13208,"journal":{"name":"IEEE Transactions on Consumer Electronics","volume":"72 3","pages":"6346-6357"},"PeriodicalIF":9.9,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148836833","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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":"Chrono-Resonant Inductive Power Symbiosis With Adaptive Neuromorphic Edge Processing for Perpetually Powered Intelligent Wearables","authors":"B. Naresh Kumar Reddy;Yellapragada Charan Krishna","doi":"10.1109/TCE.2026.3684308","DOIUrl":"https://doi.org/10.1109/TCE.2026.3684308","url":null,"abstract":"The proliferation of intelligent edge devices is fundamentally constrained by finite battery life and unpredictable energy environments. This paper introduces the Causal Reasoning-based Intelligent Power System (CRIPS) integrated with an Adaptive Neuromorphic Edge Processor (ANEP) a novel symbiotic architecture for energy-autonomous IoT nodes. Unlike conventional static systems this framework establishes a bidirectional feedback loop the ANEP’s cognitive state dynamically informs harvesting strategies while predictive energy availability dictates the processor’s operational mode. In a comprehensive 2,000-step simulation under stochastic conditions the system demonstrated 99.38% operational uptime and a <inline-formula> <tex-math>$12.2times $ </tex-math></inline-formula> energy surplus by harvesting 20,518 mJ against an operational consumption of 1,682 mJ. This metabolic abundance enabled the ANEP to operate in high-fidelity neuromorphic mode for 91.5% of the duration achieving 99.38% classification accuracy. By dynamically coupling metabolic constraints with computational goals CRIPS-ANEP bridges the gap between energy harvesting and edge intelligence, offering a robust paradigm for perpetually powered, self-aware autonomous systems.","PeriodicalId":13208,"journal":{"name":"IEEE Transactions on Consumer Electronics","volume":"72 3","pages":"6671-6680"},"PeriodicalIF":9.9,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148836834","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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":"Research on High-Efficiency Optimization and Real-Time Response of Embedded Computer Systems in Intelligent Electronic Devices","authors":"Xiao Liu;Tingting Liu;Qiangyi Cu","doi":"10.1109/TCE.2026.3668121","DOIUrl":"https://doi.org/10.1109/TCE.2026.3668121","url":null,"abstract":"The optimization of real-time performance in embedded computer systems of intelligent electronic devices presents significant challenges due to dynamic workloads, energy constraints, strict latency requirements, and operational cost considerations. Conventional static or heuristic-based methods often fail to simultaneously satisfy these competing objectives, leading to suboptimal system performance, increased energy consumption, and reduced responsiveness. To address these limitations, this study proposes a Deep Reinforcement Learning (DRL) framework that leverages agent-based strategies, including Deep Q-Networks (DQN) and Proximal Policy Optimization (PPO), to dynamically manage computational resources and adjust operational parameters in real time. The DRL agent continuously observes system states, executes actions, and learns from rewards that quantify throughput, energy efficiency, latency, and overall stability. Extensive simulation results demonstrate that the proposed DRL approach significantly outperforms conventional methods such as Energy-Aware Task Scheduling (EATS), Dynamic Voltage and Frequency Scaling (DVFS), and standard RL, achieving notable improvements in system efficiency, task processing speed, energy consumption, and reliability. By adaptively balancing performance and energy trade-offs, the DRL framework enables intelligent devices to maintain high throughput while minimizing latency and power usage, offering a robust solution for modern embedded applications operating under dynamic and constrained conditions.","PeriodicalId":13208,"journal":{"name":"IEEE Transactions on Consumer Electronics","volume":"72 3","pages":"6728-6735"},"PeriodicalIF":9.9,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148836839","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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":"Resilient Federated Vision Transformers for Alzheimer’s Disease Prediction With Brain Imaging Data Jittering","authors":"Mahyar Mohammadi;Mohammad Hossein Badiei;Mohammad Mashreghi;Keyvan Kazemi;Hamed Kebriaei","doi":"10.1109/TCE.2026.3691084","DOIUrl":"https://doi.org/10.1109/TCE.2026.3691084","url":null,"abstract":"Neural networks in Alzheimer’s MRI prediction are vulnerable to subtle adversarial perturbations and limited data diversity. In this paper, we present a robustness strategy for federated learning (FL) with vision transformers (ViTs), termed FL-ViTs, aimed at improving classification performance under domain uncertainties. We employ worst-case robust optimization techniques in FL-ViT models to effectively handle uncertainty regions and address data heterogeneity in FL. Specifically, we implement a min-max optimization framework, where the inner maximization step simulates worst-case adversarial perturbations for each local model, and the outer minimization step adjusts the model parameters to minimize loss under these conditions. Our experiments use benchmark models (e.g., ResNet-18, DenseNet121) on standard brain MRI data, applying recent robust optimization methods like jittering, TIFGSM, and DIFGSM to boost resilience and predictive accuracy. Experimentally validated, our approach achieved a 38% increase in robustness, with accuracy improving from 25.72% in the baseline scenario to 63.29% under adversarial attacks. In addition, our approach shows an efficient trade-off between robustness and accuracy, with a 0.28% improvement in accuracy, from 99.63% in the baseline to 99.90% in robust, attack-free conditions. Overall, the Jitter-robust method outperforms TIFGSM-robust and DIFGSM-robust in both attack-free and attacked scenarios, achieving the best balance between accuracy and adversarial resilience with an average accuracy of 81.34%. The code is available at <uri>https://github.com/ArgAI-Lab/Res-Fed-ViT</uri>","PeriodicalId":13208,"journal":{"name":"IEEE Transactions on Consumer Electronics","volume":"72 3","pages":"5945-5953"},"PeriodicalIF":9.9,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148836958","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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":"AI-Driven Probabilistic Day-Ahead Electricity Price Forecasting With Tail Risk Mitigation via Dynamic Spike Calibration","authors":"Tian Ni;Hui Hou;Xixiu Wu;Jiajia Yang;Zhao Yang Dong;Zhiqiang Zhang","doi":"10.1109/TCE.2026.3679105","DOIUrl":"https://doi.org/10.1109/TCE.2026.3679105","url":null,"abstract":"Accurate day-ahead electricity price forecasting remains challenging due to tail risks induced by market volatility. Conventional forecasting models often fail to capture price spikes associated with extreme price values, leading to significant tail bias. This paper presents a robust probabilistic price forecasting framework that integrates extreme value theory with deep-learning probability forecasting and post-hoc calibration. Firstly, a quantile-based sliding window is developed to identify extreme price realizations, and generalized extreme value distribution parameters are estimated using a gradient-based optimization approach for extreme value fitting. Meanwhile, a robust Box-Cox transformation and Huber loss are jointly employed to improve distributional modeling and training stability. Finally, a dynamic calibration mechanism is introduced to refine extreme price predictions and enhance the quantification of tail uncertainty. Case studies using real-world electricity market data from Guangxi, China, demonstrate the effectiveness of the proposed framework. Compared to state-of-the-art baselines, the proposed framework improves extreme-value prediction accuracy with an R2 increase of 0.46, and the average coverage errors at the 80%, 90%, and 95% confidence levels are reduced to 6.00%, 3.57%, and 2.98%, respectively. Additional experiments using Transformer and temporal convolutional network backbones further verify the strong generalization capability.","PeriodicalId":13208,"journal":{"name":"IEEE Transactions on Consumer Electronics","volume":"72 3","pages":"6417-6427"},"PeriodicalIF":9.9,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148837103","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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":"Intent-Based Multi-Agent Allocation With Nature-Inspired Online Algorithm","authors":"Minxi Feng;Nada Alzaben;Yonis Gulzar;Jiaming Pei","doi":"10.1109/TCE.2026.3668074","DOIUrl":"https://doi.org/10.1109/TCE.2026.3668074","url":null,"abstract":"This paper investigates the problem of online agent selection for intent recognition in multi-agent Intent-Based Networking (IBN) systems, where diverse intelligent agents provide intent interpretation under resource and cost constraints. To address dynamic environments, adversarial inputs, and uncertain predictions, we propose <italic>ICC-Bandit</i>, an online contextual constrained bandit framework that adaptively selects among heterogeneous agents based on intent input features, cost feedback, and predictive confidence. The term nature-inspired refers to a feedback-reinforced selection mechanism in which agent preference scores are strengthened or weakened by online performance and cost signals, analogous to reinforcement dynamics in swarm coordination, while feasibility is maintained through constraint-aware updates. The proposed method integrates constraint-aware learning with calibrated online regret minimization, ensuring bounded performance loss while respecting system-level constraints such as violation budgets and adaptation latency. We establish theoretical guarantees on dynamic regret and competitive ratio, demonstrating that ICC-Bandit achieves provably efficient learning under adversarial and stochastic conditions. Experiments on three real-world intent datasets show that ICC-Bandit consistently outperforms existing baselines in terms of regret, violation rate, and computational adaptability, achieving up to 44% reduction in dynamic regret and 14% improvement in competitive ratio.","PeriodicalId":13208,"journal":{"name":"IEEE Transactions on Consumer Electronics","volume":"72 3","pages":"6479-6487"},"PeriodicalIF":9.9,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148837117","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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":"Consumer-Centric Explainable Multimodal Digital Twin Modeling for Personalized Alzheimer's Detection in Healthcare 5.0","authors":"Hamid Asmat;Ikram Ud Din;Ahmad Almogren;Toqeer Ali Syed;Joel J. P. C. Rodrigues","doi":"10.1109/TCE.2026.3659536","DOIUrl":"https://doi.org/10.1109/TCE.2026.3659536","url":null,"abstract":"Alzheimer disease (AD), as one of the progressive neurodegenerative diseases, is recognized as one of the major causes of dementia throughout the world, and the existing methods of diagnosis and treatment prove to be highly limited. To overcome these obstacles, the proposed AD-TwinML framework integrates MRI and PET imaging with machine learning methods, which provides an effective solution to the diagnosis and management of Alzheimer. The proposed framework employes Vision transformer (ViTs), Digital twin (DT), and Time-series Transformers (TST) which will help in modeling disease progression to predict and diagnose Alzheimer’s. The experimental findings show that the framework has better accuracy (94.5%), sensitivity (91.5%), and AUC (0.946) than MM-CNN and MRI-VGG16. Moreover, AD-TwinML has lower execution time (14 seconds) and error values (6.1%), which is appropriate to use in real time clinical scenarios. The framework is consistent with the objectives of Healthcare 5.0 because it facilitates consumer-focused, individualized monitoring and proactive intervention plans.","PeriodicalId":13208,"journal":{"name":"IEEE Transactions on Consumer Electronics","volume":"72 3","pages":"6558-6565"},"PeriodicalIF":9.9,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148837050","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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":"Trustworthy Stress Detection in Consumer Wearables: Poisoning-Resilient Learning With Feature-Gated DualNet and Generative Restoration","authors":"Rashid Kamal;Aftab Ali;Ian Cleland;Chris D. Nugent;Zeineb Klai;Mohammed Jameel","doi":"10.1109/TCE.2026.3693832","DOIUrl":"https://doi.org/10.1109/TCE.2026.3693832","url":null,"abstract":"Wearable devices are increasingly used for continuous stress monitoring, yet their reliability under adversarial data manipulation remains largely unexamined. This paper presents the first systematic study of <italic>training-time poisoning</i> in consumer-grade stress detection, where adversaries perturb or inject malicious samples during model training. We frame stress recognition as a transaction-level classification problem over short, multimodal physiological windows, explicitly accounting for class imbalance, noise, and contamination. Across two widely used benchmarks (WESAD and DEAP), we show that classical machine learning models (Logistic Regression, SVM, Random Forest) exhibit severe degradation under poisoning, with Macro-F1 dropping by up to 30% and stress recall collapsing below 0.50 at 40% contamination. To address this, we introduce <italic>Feature-Gated DualNet (FG-DualNet)</i>, a dual-branch neural architecture that combines adaptive feature gating with context-preserving representation learning. We further propose a <italic>generative restoration defense</i> that leverages outlier filtering and CTGAN-based synthesis to reconstruct poisoned minority classes. This proactive defense operates at training time by identifying anomalous transactions and reconstructing stress-class distributions before model retraining. Together, these techniques sustain up to 20–25 percentage points higher recall under severe poisoning while maintaining interpretable modality-level importance (e.g., electrodermal activity and respiration). Our findings establish poisoning robustness and physiological interpretability as critical design objectives for trustworthy, stress-aware consumer electronics, and provide a reproducible pipeline for evaluating resilience in future wearable sensing systems.","PeriodicalId":13208,"journal":{"name":"IEEE Transactions on Consumer Electronics","volume":"72 3","pages":"7610-7623"},"PeriodicalIF":9.9,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11520900","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148837153","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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":"Secure CNN Inference With Adaptive Activation Functions Based on Mutual Information Optimization","authors":"Lingling Wu;Xu An Wang;Wenhao Liu;Haibo Lei;Wei Zhao;Jianping Zhang;Ming Lv","doi":"10.1109/TCE.2026.3669387","DOIUrl":"https://doi.org/10.1109/TCE.2026.3669387","url":null,"abstract":"As an increasing number of neural network models are deployed in the cloud, privacy concerns have surged, primarily due to the exposure of plaintext user data. To address the core challenge in ciphertext inference where polynomial activation functions struggle to balance model expressive capability and encrypted computation efficiency, this paper proposes a ciphertext CNN inference scheme with mutual information-optimized adaptive polynomial activation functions. Based on the CKKS homomorphic encryption algorithm, the scheme quantifies the correlation between features of each neural network layer and labels via mutual information, dynamically matching the degree and coefficients of polynomial activation functions. First, in the plaintext training phase, mutual information statistics are utilized to classify the functional importance levels of each network layer. Second, polynomial coefficients are optimized through end-to-end training, and combined with threshold filtering and numerical scaling in the ciphertext domain, ensuring the coefficients comply with the constraints of encrypted computation. Finally, a lightweight CNN architecture is integrated with the TenSEAL encryption framework to achieve end-to-end ciphertext inference. Experiments on the MNIST dataset and Fashion-MNIST dataset demonstrate that the proposed scheme achieves a plaintext model accuracy of 98.34%, 95.62% respectively and a ciphertext inference accuracy of 98.13%, 95.31% respectively. It significantly outperforms fixed-degree polynomial activation schemes, realizing the synergistic optimization of privacy protection, inference accuracy, and computational efficiency under a 128-bit security level.","PeriodicalId":13208,"journal":{"name":"IEEE Transactions on Consumer Electronics","volume":"72 3","pages":"6771-6786"},"PeriodicalIF":9.9,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148837260","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}