×
The submission system is temporarily under maintenance. Please send your manuscripts to
Go to Editorial ManagerWith the rapid popularization of 5G edge communication systems, their decentralized architecture and the ultra-large attack surface brought by massive access devices have created severe network security challenges. Although current artificial intelligence-based intrusion detection systems and conventional federated learning schemes can achieve high intrusion detection accuracy, they universally suffer from two major limitations: excessive computational overhead and vulnerability to poisoning attacks, which make them unable to adapt to resource-constrained edge deployment environments. To address this issue, this paper proposes a lightweight anti-poisoning federated learning framework for low-latency secure 5G edge communications, with two core design elements: First, it integrates lightweight GRU/CNN models and a trust-aware aggregation mechanism to weaken the negative impact of malicious model updates; second, it introduces a deviation-based poisoning detection strategy to identify and filter adversarial updates during the federated learning process. We used the public CICIDS2017 dataset to complete tests in a simulated 5G edge environment equipped with 100 to 1000 edge devices, covering multiple scenarios including DDoS, MITM, and poisoning attacks. The results show that compared with existing similar schemes, the proposed framework reaches a maximum detection accuracy of 98.3\%, reduces communication latency by 30-35\%, cuts computational overhead by 40-60\%, and sees only a negligible drop in accuracy under poisoning attacks. It can support a wide range of real-time 5G edge applications such as intelligent transportation, smart cities, and the industrial Internet of Things.
Harsh industrial environments such as oilfields present unique challenges to electronic systems, including extreme temperatures, limited connectivity, power constraints, and operational unpredictability. Traditional Internet of Things (IoT) deployments often fail to adapt in real-time, exposing systems to risks such as data loss, late anomaly detection, or critical failure. This paper proposes a lightweight, Artificial Intelligence (AI)-driven eSystem architecture tailored for such conditions, integrating edge intelligence, secure communication, and self-adaptive mechanisms. We demonstrate the framework's viability through simulating a case study of real-time sensor data from pipeline infrastructure, applying a Long Short-Term Memory (LSTM)-based anomaly detection model deployed at the edge. Results show significant improvements in detection latency, bandwidth efficiency, and system resilience. The framework offers a modular blueprint for deploying AI-enhanced eSystems across energy, mining, and remote critical infrastructure domains.
As Internet of Things (IoT) devices continue to spread, they also create many new entry points for cyberattacks. Traditional security methods struggle to keep up, which makes smarter and more adaptive defenses necessary. This paper introduces an Artificial Intelligence (AI)–driven threat intelligence framework designed to improve intrusion detection in diverse IoT networks. The framework combines Machine Learning (ML) and Deep Learning (DL) models to detect malicious activity more accurately across different types of network traffic. To evaluate the approach, three widely used benchmark datasets—UNSW-NB15, CIC-IDS2017, and IoT-Botnet—were used. Experimental results show that the proposed hybrid Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM) model performs very well. It achieved 97% accuracy, a 0.95 F1-score, and a 0.98 Receiver Operating Characteristic – Area Under the Curve (ROC-AUC) on the UNSW-NB15 dataset, outperforming traditional ML models such as Random Forest, which reached 94% accuracy. While DL models provided better detection performance and stronger generalization, ML models proved to be much faster, with nearly three times lower inference latency—about 3 milliseconds per network flow. This makes them more suitable for real-time deployment at the IoT edge, where computing resources are limited. Overall, the proposed hybrid approach strikes a practical balance between detection accuracy and processing speed, offering a scalable and robust foundation for AI-based IoT threat intelligence in real-world environments.
Keeping IoT networks running reliably in harsh environments is still a tough problem. Sensors wear out, communication links are unreliable, and maintenance quickly becomes expensive. These issues make traditional monitoring approaches fragile and slow to react. This work presents a self-adaptive, AI-driven Digital Twin framework that continuously tracks the real state of an IoT network and flags failures before they actually happen. The system mirrors the physical network in real time by combining edge-level data preprocessing, physics-aware Digital Twin simulations, and well-chosen deep learning models for anomaly detection and remaining useful life estimation. To test the idea, we simulated a network of 50 IoT nodes operating under realistic harsh conditions, including thermal stress, high humidity, and signal interference. The results are hard to ignore. The proposed framework reached 91% prediction accuracy, detected problems 27 seconds earlier on average, and improved overall network reliability from 84% to 96% compared to standard threshold-based monitoring. The takeaway is straightforward: pairing AI analytics with Digital Twin technology enables proactive and resilient IoT operation in environments where conventional monitoring quickly falls apart. This work lays a practical foundation for deploying AI-enhanced Digital Twins in real-world, next-generation IoT systems, where reliability actually matters.
The combination of the Internet of Things (IoT) and Software-Defined Networking (SDN) has quickly revolutionized the world of industry, and also brought with it some severe security risks. Despite its recent success, Deep learning (DL) can be a powerful method for Network Intrusion Detection systems (NIDS), but often the computational and memory requirements are too high for edge devices. In this paper, a collaborative optimization framework with lightweight deep-learning IDS models suitable for SDN-IoT environments is proposed. We formalize and test three collaborative compression schemes based on combining pruning, clustering, and quantization, yet maintaining the structural advantages of each stage: Sparsity-Preserving Quantization-Aware Training (PQAT), Cluster-Preserving Quantization-Aware Training (CQAT), and Sparsity-and-Cluster-Preserving Quantization-Aware Training (PCQAT). The three schemes are evaluated on four canonical architectures (LSTM, RNN, DNN and CNN) on the proposed SICA dataset with respect to DDoS, Probe and Brute-Force (BFA).The results show that aggressive single-stage compression such as clustering can shrink model size by up to 94%, but at the cost of predictive confidence; PCQAT, by contrast, acts as a stabilizer that recovers this lost confidence. In particular, CNN-PCQAT and DNN-PCQAT deliver the most favorable trade-off, sustaining high detection accuracy on all three attack classes with inference latency below 0.01 ms. The framework therefore offers a practical, deployment-oriented guideline for implementing real-time IDS on resource-constrained SDN-IoT edge devices.
Predictive maintenance (PdM) is a key enabler of reliability and cost reduction in Industrial Internet of Things (IIoT) systems. However, conventional centralized learning approaches require raw data aggregation, raising serious concerns regarding data privacy, communication overhead, and single points of failure. Federated Learning (FL) addresses part of this challenge by enabling distributed model training, yet centralized FL architectures still rely on a central aggregator. This paper proposes a decentralized federated learning (DFL) framework based on a ring topology for multi-class predictive maintenance in IIoT environments. The proposed approach enables collaborative model training among distributed edge nodes without relying on a central server, thereby enhancing system robustness, scalability, and data privacy. A lightweight deep neural network (DNN) is employed to classify five machine operating states using multi-sensor data, including vibration, temperature, and current measurements. Extensive experiments are conducted using a real-world sensor dataset collected from three-phase AC motors under multiple fault conditions. The performance of the proposed DFL approach is systematically compared against a non-federated baseline and a centralized FL (CFL) scheme. Experimental results demonstrate that the decentralized framework achieves clas-sification performance comparable to centralized FL, with an accuracy of up to 99.83%, while eliminating the dependency on a central coordinator. The findings confirm the feasibility of decentralized federated learning for predictive maintenance applica-tions, particularly in distributed industrial settings where data privacy, fault tolerance, and communication efficiency are critical.