Transactions on Emerging Telecommunications Technologies最新文献

This investigation delineates a dual-domain computational paradigm integrating machine learning (ML)-based parametrization of gaseous optical properties with ensemble-driven anomaly detection in Vehicular Ad Hoc Networks (VANETs), aimed at accelerating predictive fidelity in climate-cyber-physical simulations. The study formulates a Random Forest (RF) surrogate model to approximate nonlinear radiative transfer coefficients and simultaneously classify spatiotemporally correlated anomalies induced by cyber-physical perturbations within VANET topologies. A synthetically constructed dataset comprising 500 multi-dimensional vehicular communication profiles encoding instantaneous velocity vectors, geospatial coordinates, signal-to-noise ratios, inter-vehicular latency, and communication fidelity metrics was employed for rigorous model training and validation. The surrogate framework achieved a coefficient of determination ($$) of 0.963, Mean Squared Error (MSE) of $$, and Mean Absolute Error (MAE) of 0.028, reflecting sub-3% deviation from empirically derived ground truth in both anomaly classification and optical property approximation. Notably, latency analysis indicated a 35% reduction relative to conventional rule-based intrusion detection paradigms, underscoring the potential of ensemble-based surrogate learning for real-time, computationally efficient detection of VANET anomalies under high-dimensional, stochastic environments. Prospective enhancements include hybridization with temporal convolution networks and federated learning strategies to enable edge-resident deployment while preserving high-resolution optical and cyber-physical predictive accuracy. This study substantively demonstrates the feasibility of co-optimized surrogate modeling frameworks for synergistic acceleration of climate simulation fidelity and cyber-physical system resilience.

Vehicular Ad Hoc Networks (VANETs), a crucial part of Intelligent Transportation Systems, provide real-time vehicle-RSU communication. VANETs are subject to DDoS assaults, which may undermine safety-critical systems due to their dynamic and dispersed nature. VANET DDoS detection solutions frequently use centralized designs that need raw data aggregation (DA), which causes scalability, communication cost, and privacy difficulties. Additionally, traditional models fail to adapt to vehicular surroundings' varied and frequently changing traffic patterns. Federated Anomaly Detection with Personalized Autoencoders (FAD-PAE) is proposed to address these issues. Vehicles and roadside equipment train lightweight Autoencoders (AE) locally to simulate regular traffic behavior, exchanging just model updates via federated learning. Personalized fine-tuning at each node adapts to local traffic changes, while safe and strong aggregation protects against compromised clients. The VANET-based cooperative and privacy-preserving DDoS detection approach reduces false alarms and communication costs. Experimental results show superior detection accuracy (ACC), reaction speed, and flexibility compared to centralized techniques.

The rapid expansion of IoT ecosystems has intensified challenges in securely managing large volumes of sensitive data generated by devices operating under constraints of limited computation, energy, and bandwidth. Conventional key management and authentication mechanisms struggle with scalability and efficiency, increasing susceptibility to emerging security threats. To address these limitations, this study proposes a lightweight, post-quantum secure key management framework that integrates lattice-based encryption with Ciphertext-Policy Attribute-Based Encryption (CP-ABE). The model incorporates SHA-512–based authentication to ensure device integrity and employs Elastic Search with MapReduce for scalable data compression and handling. End-to-end confidentiality is maintained through secure encryption, dynamic re-encryption, and fine-grained cloud-based access control. MATLAB simulations validate system performance, demonstrating improved encryption and decryption times of 145 and 160 ms, surpassing traditional CP-ABE and RSA approaches. Key generation and update times are reduced to 120 and 95 ms, enabling dynamic policy management. Authentication accuracy reaches 98.7% precision and 97.9% recall, ensuring reliable device verification, while Elastic Search and MapReduce achieve 250 MB/s throughput and reduce network load by 35% via data-locality scheduling. Re-encryption overhead remains low at 130 ms, with minimal resource usage (20% CPU, 25 MB memory), supporting constrained devices. Scalability tests show only modest latency growth up to 200 nodes. The proposed lattice-based CP-ABE framework offers a secure, efficient, and scalable solution for next-generation IoT environments. Future work will extend applicability to larger networks and integrate AI-driven adaptive security for enhanced resilience.

Object detection has gradually developed into a popular research area because of the rife use of Remote Sensing Images (RSIs) in military and civil domains. However, complex backgrounds and problems with small items are the two biggest obstacles in object detection. Deep learning techniques have a significant advantage for object detection over conventional methods that rely on manually derived characteristics, and they have a lot of attention. Due to the wide range of objects and complex visual backgrounds in RSIs, current deep-learning algorithms in the area of RSI object detection leave much to be desired. For this case, algorithms need to be specifically optimized. In this paper, a novel optimization-driven Enhanced Faster Region Convolutional Neural Network (FRCNN) is proposed for object detection. This approach has five modules, namely pre-processing, data augmentation, segmentation, feature extraction, and object detection. The detection process is performed using Enhanced FRCNN and it is trained by the proposed Gazelle White Shark Optimization Algorithm (GWSOA). Extensive experiments in high-resolution RSI data sets have exposed the efficacy of the proposed approach. The novel approach achieved better accuracy of 97.31%, Mean Average Precision (MAP) of 97.85%, precision of 97.76%, recall of 96.16%, F-score of 95.78%, and error rate of 2.69.

The rapid emergence of Internet of Things (IoT) applications such as smart cities, intelligent transportation, healthcare, logistics, and wearable systems has increased the demand for scalable and low-latency computing infrastructure. Despite the widespread adoption of traditional cloud computing platforms to handle these applications, offloading all tasks to centralized cloud data centers results in significant latency, high energy consumption, and increased operational costs. Many of these applications are delay-sensitive and computationally intensive, requiring immediate decision-making and localized processing. To address these limitations, we propose a prioritized constraint-aware task offloading mechanism (PCATOM) that efficiently schedules and offloads tasks by considering task-level and resource-level constraints in a fog-cloud computing environment. PCATOM leverages the Deep Deterministic Policy Gradient (DDPG) algorithm, a policy gradient reinforcement learning method designed to balance exploration and exploitation while dynamically learning optimal offloading strategies. The framework reduces overall latency, energy consumption, and execution cost by making intelligent decisions about where and how to offload tasks. PCATOM was implemented using the SimPy simulation framework and evaluated using a combination of statistical workloads and real-world parallel computing traces from NASA and HPC2N. Experimental results show that PCATOM consistently outperforms baseline models such as DQN and A2C, achieving up to 32.5% lower latency, 28.7% lower energy consumption, and 18.4% higher throughput. These results demonstrate the effectiveness and scalability of PCATOM in dynamic and diverse fog-cloud environments.

The terahertz (THz) range is perfect for high-throughput applications like enhanced sensing, ultra-fast wireless backhauls, and real-time virtual and augmented reality because of its availability of unoccupied spectrum, which permits unparalleled bandwidths. The novel THz designs and components are being propelled by recent developments in nanomaterials, photonic devices, and intelligent surfaces. Additionally, the integration of THz waves with sensing and security capabilities opens up new paradigms in secure communication, healthcare applications, and intelligent settings. To achieve reliable and extensible THz communication systems, this article delves into the necessary technologies, propagation properties, and future study paths. Photonic crystals (PhC) have recently gained popularity for use in cutting-edge electromagnetic fields because of their remarkable controllability over the transmission of electromagnetic waves. The use of PhC ideas along with old and new antenna designs offers a way to create antennas that are very directional, slim, and adjustable. This article proposes a new design that combines the Kagome lattice PhC structure with a modified patch antenna that uses a Gingham sequence. The air holes in the Kagome lattice are optimized to enhance antenna characteristics like return loss (RL), gain, voltage standing wave ratio (VSWR), directivity, and so on. The proposed Kagome lattice PhC antenna works at THz frequency; hence, it is suitable for medical applications like cancer detection, and so on.

Aiming at the problem that ADS-B (Automatic Dependent Surveillance Broadcast) data jump characteristics and multi-source heterogeneous errors are difficult to dynamically associate, this paper constructs a data mining framework that combines time series anomaly detection and multi-modal feature fusion. First, this paper performs missing value interpolation, time alignment and standardization preprocessing on the collected ADS-B data, energy status parameters, material environment parameters and flight environment information; secondly, based on the LSTM-AE (Long Short-Term Memory-Autoencoder) model, an unsupervised model for ADS-B data jump detection is constructed, which identifies the jump point by reconstructing the error and extracts jump features with physical significance. Finally, this paper introduces the Transformer architecture, cross-modally fuses the jump features with multi-source heterogeneous parameters, constructs a semantic alignment mechanism, and combines the error evolution modeling strategy to explore the complex influence mechanism of different factors on positioning errors. Experimental results show that the proposed method has better jump recognition accuracy (average 90.7%) and error prediction mean square error (average 5.81) than Isolation Forest and Variational Autoencoder (VAE), providing a new technical path for improving the positioning reliability and safety control of unmanned aerial vehicles (UAVs).

The traditional cybersecurity solution cannot offer the necessary protection against all forms of cyberattacks for management systems since cyberattacks are of a scattered nature. Thus, an efficient security method is suggested in this work to prevent data loss from the information system using optimization and deep learning. The major aim of this work is to offer suitable recommendations for developing a cybersecurity decision support model for information systems with risk analysis and cybersecurity models. In the information system, the security threats are eliminated using this model. So, the authentication, prediction of the risk levels, and the mitigation of the security threats are regarded as the major objectives of this research work. From the CICIDS 2017 KNN dataset, UNSW-NB15 dataset, and NSL-KDD dataset, the data is collected. The collected data is given to the executed Deep Security Network (DSecNet). In the DSecNet, the Restricted Boltzmann Machine (RBM) is utilized to retrieve the features and then the extracted features from the RBM are fused with optimized weights, which are selected using the Enhanced Sandpiper Optimization Algorithm (ESOA). The obtained weighted features are further provided to the Deep Capsule Neural Network (DeepCapsNet). The implemented DSecNet is used for verifying the authentication of the user, detecting the intrusion, and determining the associated risk levels caused by the attacks on the information system. Once the threats are detected, the mitigation of the threat based on their risk level is carried out. The effectiveness of the proposed model is proven by the validation process. From the simulation outcomes, the accuracy rate of the developed model is 97.11% in the NSL-KDD dataset and 96.70% in the UNSW-NB15 dataset based on intrusion detection performance. Therefore, the efficacy of the developed security system is higher for the performance of the authentication, intrusion detection, and risk prediction, which elevates the security of the cyber networks.