Internet Technology Letters最新文献

Industrial Internet of Things (IIoT) systems generate massive behavioral data, demanding efficient analytics frameworks for real-time monitoring. This study proposes a multi-dimensional feature fusion framework integrating XGBoost, tailored for IIoT-driven behavioral pattern recognition. A four-dimensional architecture is constructed to analyze critical attributes across contact degree, status, duration, and social relations, leveraging edge-computed IIoT footprints (e.g., mobile signaling, network interaction data). The framework defines three behavioral modes and achieves 98.89% precision, 98.85% recall, and 98.85% F1-score via XGBoost. Feature importance analysis identifies key indicators such as mobile number status and interaction frequency. This work demonstrates the potential of harmonizing AI with IIoT data fusion, providing a scalable solution for real-time monitoring in Industrial Internet and future network architectures.

The accuracy of existing ground wire position identification devices is often compromised due to electromagnetic interference. To address this issue, this paper proposes an intelligent recognition device that integrates radio frequency identification (RFID) technology with ultra-wideband (UWB) positioning. By leveraging the broad spectrum of UWB signals to disperse energy, the system effectively reduces electromagnetic interference and enhances positioning accuracy. Experimental results demonstrate that the positioning error remains consistently within the range of 0.06–0.1 m. Even in environments with significant electromagnetic interference, the increase in error is minimal, indicating strong anti-interference performance. The device proposed in this paper, which utilizes RFID-UWB technology, significantly improves the precision of ground wire positioning and offers robust technical support for intelligent identification tasks requiring high accuracy.

This work presents a Predictive Large Model-Driven Framework (PLMF) for Wireless English Education Platforms (WEEPs) that integrates real-time Quality of Experience (QoE) forecasting, CEFR-aware semantic simplification, and adaptive content delivery in a unified, feedback-driven architecture. To support system evaluation, we construct EduQoE-PLMF, a multimodal dataset comprising CEFR-tagged content, simulated network traces, behavioral logs, and user-rated QoE labels. PLMF is benchmarked against five representative baselines across three key tasks. Experimental results show that PLMF achieves superior performance in QoE prediction (MSE: 0.025, R²: 0.89), content simplification (SARI: 44.9, Readability: 2.9), and learner engagement (TCR: 83.2%, DR: 11.4%, SSS: 4.3). Ablation studies and heatmap analysis further reveal the complementary value of each system module. These findings demonstrate the effectiveness of combining predictive reasoning, personalization, and delivery optimization to enable robust and learner-centered wireless education systems.

Data aggregation is one major problem in heterogeneous wireless sensor networks (WSNs) where nodes possess varying sensing, computation, and communication capabilities. In order to fulfill the requirements of energy efficiency, latency, and optimization of the network lifetime, we introduce the QIBOA_HHO_Hybrid protocol, which is a mix of the Quantum-Inspired Binary Optimization Algorithm (QIBOA) and the Harris Hawks Optimization (HHO) algorithm. The hybrid protocol synergistically blends QIBOA's quantum-inspired parallel search to gain faster convergence with HHO's adaptive exploitation methods to optimize routing and clustering decisions dynamically. By prioritizing the most important energy-aware cluster head (CH) selection based on proximity and residual energy, the protocol balances the load and minimizes energy consumption. Simulation results verify that QIBOA_HHO_Hybrid outperforms conventional protocols SEP, DEEC, Z-SEP, and PSO-ECSM, with less latency, more throughput, and more network lifetime. By fusing quantum optimization while simulations suggest a compromise with energy efficiency and latency compared to some existing protocols, adaptive clustering, and HHO's cooperative predation-inspired methods, scalability and reliability are enhanced in dynamic environments, and it is a trusted solution to large-scale heterogeneous WSNs.

Beamforming and channel estimation are fundamental components of 5G massive MIMO (multiple-input–multiple-output) systems, particularly in the millimeter-wave (mmWave) spectrum, where high-frequency transmissions are susceptible to path loss and signal degradation. The growing demand for ultrareliable low-latency communication (URLLC) and high-quality services necessitates advanced, adaptive techniques to manage the highly dynamic nature of mmWave channels. This study proposes a novel framework that integrates dual-attention convolutional neural networks (DSCN-PAN) with reformed poplar optimization (RePO) to enhance beamforming accuracy and channel estimation efficiency in 5G massive MIMO systems. Compared to conventional methods, the proposed model demonstrates significant performance gains, including over 90% improvement in spectral efficiency, 99.41% beam alignment precision, a 99.5% enhancement in Channel State Information (CSI) estimation, and a 99.2% reduction in bit error rate (BER). The DSCN-PAN-RePO architecture effectively supports dynamic and complex communication environments, offering a scalable and energy-efficient solution for next-generation wireless networks.

In this article, we introduce the reconfigurable intelligent surfaces (RISs) restrict their general adoption, which employs digital pre-distortion, deep learning-based correction, and adaptive filtering to counteract real-time RF impairments. The technique is highly applicable to future 6G networks because it enhances MIMO performance by reducing BER, improving phase noise resilience, and achieving the highest spectral efficiency. Thermal noise, phase noise, and nonlinearity loss are RF impairments that significantly reduce the effectiveness of MIMO communication in 6G networks. Signal distortion, phase instability, and spectrum inefficiencies are the consequences of these impairments, which further increase BER and reduce capacity. A dynamic distortion mitigation framework is required because conventional compensating strategies cannot respond to new scenarios in real time. These approaches come with extra latency and power usage, making them less suitable for real-time use in 6G. However, there remains a challenge to using ML-based adaptive filtering on high-speed and low-power hardware, even though it has been at the forefront regarding dynamically compensating RF impairments. The cost and complexity of deployment of hybrid beamforming and reconfigurable intelligent surfaces (RISs) restrict their general adoption, yet they enhance MIMO performance in RF impairment. The basic challenge for smooth operation in 6G-enabled MIMO systems is to develop adaptive, low-power, and computationally efficient solutions.

Advanced Persistent Threat (APT) attacks pose severe security risks to mobile edge networks due to their stealthy, long-term, and multi-stage nature. This paper proposes MERA-RD, a novel real-time APT detection framework that integrates multi-source data fusion, a Spatio-Temporal Graph Neural Network (ST-GNN) for temporal–spatial correlation modeling, and a Deep Q-Network (DQN)-based adaptive threshold adjustment mechanism. The framework is designed to address the challenges of heterogeneous device environments, dynamic traffic patterns, and stringent latency constraints in Mobile Edge Computing scenarios. Experimental evaluations in both simulated and real-world environments demonstrate that MERA-RD achieves high detection accuracy with low latency, validating its potential for practical deployment. The proposed approach provides a promising solution for enhancing the security of edge-based intelligent systems in the era of 6G networks.

With the rapid development of the Internet of Things (IoT), the demand for personalized fashion design in the IoT environment is growing, and fashion recommendation has gradually become a new research hotspot. However, existing fashion recommendation methods are often designed based on a single modality and contain a large number of parameters, making them unable to be effectively deployed on IoT edge devices with limited computing ability. Inspired by this, this paper proposes a novel personalized fashion color recommendation (FashionCR) framework based on a lightweight large vision-language model for fashion design in the IoT environment. Specifically, this framework consists of an IoT-based fashion color recommendation system and the FashionCR model. The recommendation system mainly introduces how to train the FashionCR model and deploy it to the edge devices. The FashionCR model leverages the visual branch of the CLIP model to accurately learn the physiological features of different individuals, such as skin color and face shape, and utilizes the text branch to efficiently process the text intentions input by users. Meanwhile, in order to meet the limited resources of the IoT environment, a lightweight modification has been implemented to the CLIP model. In addition, the 4-season color theory is integrated into the FashionCR framework to achieve accurate color recommendation. Experimental results show that this framework performs excellently in various metrics, providing a new solution for the field of fashion design in the IoT environment and effectively improving the accuracy and personalization of color recommendation.

This manuscript provides a smoothing technique for objective penalty functions in inequality-constrained optimization problems. A non-smooth penalty function is defined which is subjected to a new smoothing technique to make it smooth. The error estimates for the original and the smoothed problem are discussed. A procedure is illustrated for the development of the solution of the inequality-constrained optimization problem and is shown to be convergent under certain specified conditions. The same can be incorporated in various application areas like Wireless Sensor Networks in the form of giving penalties to sensor nodes not fulfilling the network performance criteria and also in some other aspects like 5G communication.