Search Results (48)

Accurate and real-time classification of low-latency Internet traffic is critical for applications such as video conferencing, online gaming, financial trading, and autonomous systems, where millisecond-level delays can degrade user experience. Existing methods for low-latency traffic classification, reliant on raw temporal features or static statistical analyses, fail to capture dynamic frequency patterns inherent to real-time applications. These limitations hinder accurate resource allocation in heterogeneous networks. This paper proposes a novel framework integrating wavelet transform (WT) and artificial neural networks (ANNs) to address this gap. Unlike prior works, we systematically apply WT to commonly used temporal features—such as throughput, slope, ratio, and moving averages—transforming them into frequency-domain representations. This approach reveals hidden multi-scale patterns in low-latency traffic, akin to structured noise in signal processing, which traditional time-domain analyses often overlook. These wavelet-enhanced features train a multilayer perceptron (MLP) ANN, enabling dual-domain (time–frequency) analysis. We evaluate our approach on a dataset comprising FTP, video streaming, and low-latency traffic, including mixed scenarios with up to four concurrent traffic types. Experiments demonstrate 99.56% accuracy in distinguishing low-latency traffic (e.g., video conferencing) from FTP and streaming, outperforming k-NN, CNNs, and LSTMs. Notably, our method eliminates reliance on deep packet inspection (DPI), offering ISPs a privacy-preserving and scalable solution for prioritizing time-sensitive traffic. In mixed-traffic scenarios, the model achieves 74.2–92.8% accuracy, offering ISPs a scalable solution for prioritizing time-sensitive traffic without deep packet inspection. By bridging signal processing and deep learning, this work advances efficient bandwidth allocation and enables Internet Service Providers to prioritize time-sensitive flows without deep packet inspection, improving quality of service in heterogeneous network environments. Full article

With the rapid growth of urban transportation demand, traditional taxi systems face challenges such as supply–demand imbalances and low dispatch efficiency. These methods, which rely on static data and predefined strategies, struggle to adapt to dynamic traffic environments. To address these issues, this paper proposes a dual-layer Taxi Dispatch and Empty-Vehicle Repositioning (TDEVR) optimization framework based on Multi-Agent Reinforcement Learning (MARL). The framework separates the tasks of taxi matching and repositioning, enabling efficient coordination between the decision-making and execution layers. This design allows for the real-time integration of both global and local supply–demand information, ensuring adaptability to complex urban traffic conditions. A Multi-Agent Dual-Attention Reinforcement Learning (MADARL) algorithm is proposed to enhance decision-making and coordination, combining local and global attention mechanisms to improve local agents’ decision-making while optimizing global resource allocation. Experiments using a real-world New York City taxi dataset show that the TDEVR framework with MADARL leads to an average improvement of 20.63% in the Order Response Rate (ORR), a 15.29 increase in Platform Cumulative Revenue (PCR), and a 22.07 improvement in the Composite Index (CI). These results highlight the significant performance improvements achieved by the proposed framework in dynamic scenarios, demonstrating its ability to efficiently adapt to real-time fluctuations in supply and demand within urban traffic environments. Full article

The rapid development of Network Function Virtualization (NFV) enables network operators to deliver customized end-to-end services through Service Function Chains (SFCs). However, existing two-stage deployment strategies fail to jointly optimize the placement of Virtual Network Functions (VNFs) and the routing of service traffic, resulting in inefficient resource utilization and increased service latency. This study addresses the challenge of maximizing the acceptance rate of service requests under resource constraints and latency requirements. We propose DRL-FJM, a novel dynamic SFC joint mapping orchestration algorithm based on Deep Reinforcement Learning (DRL). By holistically evaluating network resource states, the algorithm jointly optimizes node and link mapping schemes to effectively tackle the dual challenges of resource limitations and latency constraints in long-term SFC orchestration scenarios. Simulation results demonstrate that compared with existing methods, DRL-FJM improves total traffic served by up to 42.6%, node resource utilization by 17.3%, and link resource utilization by 26.6%, while achieving nearly 100% SFC deployment success. Moreover, our analysis reveals that the proposed algorithm demonstrates strong adaptability and robustness under diverse network conditions. Full article

The increasing deployment of autonomous vehicles (AVs) has exposed critical vulnerabilities in traffic sign classification systems, particularly against adversarial attacks that can compromise safety. This study proposes a dual-purpose defense framework based on convolutional autoencoders to enhance robustness against two prominent white-box attacks: Fast Gradient Sign Method (FGSM) and Projected Gradient Descent (PGD). Experiments on the German Traffic Sign Recognition Benchmark (GTSRB) dataset show that, although these attacks can significantly degrade system performance, the proposed models are capable of partially recovering lost accuracy. Notably, the defense demonstrates strong capabilities in both detecting and reconstructing manipulated traffic signs, even under low-perturbation scenarios. Additionally, a feature-based autoencoder is introduced, which—despite a high false positive rate—achieves perfect detection in critical conditions, a tradeoff considered acceptable in safety-critical contexts. These results highlight the potential of autoencoder-based architectures as a foundation for resilient AV perception while underscoring the need for hybrid models integrating visual-language frameworks for real-time, fail-safe operation. Full article

To address energy management challenges for intelligent connected automated range-extended electric vehicles under vehicle-road cooperative environments, a hierarchical energy management strategy (EMS) based on speed prediction is proposed from the perspective of multi-objective optimization (MOO), with comprehensive system performance being significantly enhanced. Focusing on connected car-following scenarios, acceleration sequence prediction is performed based on Kalman filtering and preceding vehicle acceleration. A dual-layer optimization strategy is subsequently developed: in the upper layer, optimal speed curves are planned based on road network topology and preceding vehicle trajectories, while in the lower layer, coordinated multi-power source allocation is achieved through EMS_MPC-P, a Bayesian-optimized model predictive EMS based on Pontryagin’ s minimum principle (PMP). A MOO model is ultimately formulated to enhance comprehensive system performance. Simulation and bench test results demonstrate that with SoC₀ = 0.4, 7.69% and 5.13% improvement in fuel economy is achieved by EMS_MPC-P compared to the charge depleting-charge sustaining (CD-CS) method and the charge depleting-blend (CD-Blend) method. Travel time reductions of 62.2% and 58.7% are observed versus CD-CS and CD-Blend. Battery lifespan degradation is mitigated by 16.18% and 5.89% relative to CD-CS and CD-Blend, demonstrating the method’s marked advantages in improving traffic efficiency, safety, battery life maintenance, and fuel economy. This study not only establishes a technical paradigm with theoretical depth and engineering applicability for EMS, but also quantitatively reveals intrinsic mechanisms underlying long-term prediction accuracy enhancement through data analysis, providing critical guidance for future vehicle–road–cloud collaborative system development. Full article

Vehicle detection algorithms constitute a fundamental pillar in intelligent driving systems and smart transportation infrastructure. Nevertheless, the inherent complexity and dynamic variability of traffic scenarios present substantial technical barriers to robust vehicle detection. While visual transformer-based detection architectures have demonstrated performance breakthroughs through enhanced perceptual capabilities, establishing themselves as the dominant paradigm in this domain, their practical implementation faces critical challenges due to the quadratic computational complexity inherent in the self-attention mechanism, which imposes prohibitive computational overhead. To address these limitations, this study introduces Mamba RT-DETR (MRD), an optimized architecture featuring three principal innovations: (1) We devise an efficient vehicle detection Mamba (EVDMamba) network that strategically integrates a linear-complexity state space model (SSM) to substantially mitigate computational overhead while preserving feature extraction efficacy. (2) To counteract the constrained receptive fields and suboptimal spatial localization associated with conventional SSM sequence modeling, we implement a multi-branch collaborative learning framework that synergistically optimizes channel dimension processing, thereby augmenting the model’s capacity to capture critical spatial dependencies. (3) Comprehensive evaluations on the BDD100K benchmark demonstrate that MRD architecture achieves a 3.1% enhancement in mean average precision (mAP) relative to state-of-the-art RT-DETR variants, while concurrently reducing parameter count by 55.7%—a dual optimization of accuracy and efficiency. Full article

With the rapid proliferation of electric vehicles (EVs), urban logistics faces increasing challenges in optimizing vehicle routing. This paper presents a new modeling framework for the Electric Vehicle Routing Problem (EVRP), where multiple electric trucks serve a set of customers within their capacity limits. The model incorporates critical EV-specific constraints, including limited battery range, charging demand, and dynamic urban traffic conditions, with the objective of minimizing total delivery cost. To efficiently solve this problem, a Dual Population Cooperative Genetic Algorithm (DPCGA) is proposed. The algorithm employs a dual-population mechanism for global exploration, effectively expanding the search space and accelerating convergence. It then introduces local refinement operators to improve solution quality and enhance population diversity. A large number of experimental results demonstrate that DPCGA significantly outperforms traditional algorithms in terms of performance, achieving an average 3% improvement in customer satisfaction and a 15% reduction in computation time. Furthermore, this algorithm shows superior solution quality and robustness compared to the AVNS and ESA-VRPO algorithms, particularly in complex scenarios such as adjustments in charging station layouts and fluctuations in vehicle range. Sensitivity analysis further verifies the stability and practicality of DPCGA in real-world urban delivery environments. Full article

With the rapid acceleration of global urbanization and the advent of smart city initiatives, large metropolises confront the dual challenges of surging logistics demand and constrained surface transportation resources. Traditional surface logistics networks struggle to support sustainable urban development in high-density areas due to traffic congestion, high carbon emissions, and inefficient last-mile delivery. This paper addresses the layout optimization of a hub-and-spoke underground space logistics system (ULS) network for smart cities under stochastic scenarios by proposing an immune-inspired multi-objective particle swarm optimization (IS-MPSO) algorithm. By integrating a stochastic robust Capacity–Location–Allocation–Routing (CLAR) model, the approach concurrently minimizes construction costs, maximizes operational efficiency, and enhances underground corridor load rates while embedding probability density functions to capture multidimensional uncertainty parameters. Case studies in Beijing’s Fifth Ring area demonstrate that the IS-MPSO algorithm reduces the total objective function value from 9.8 million to 3.4 million within 500 iterations, achieving stable convergence in an average of 280 iterations. The optimized ULS network adopts a “ring–synapse” topology, elevating the underground corridor load rate to 59% and achieving a road freight alleviation rate (RFAR) of 98.1%, thereby shortening the last-mile delivery distance to 1.1 km. This research offers a decision-making paradigm that balances economic efficiency and robustness for the planning of underground logistics space in smart cities, contributing to the sustainable urban development of high-density regions and validating the algorithm’s effectiveness in large-scale combinatorial optimization problems. Full article

Traffic sign classification (TSC) based on deep neural networks (DNNs) plays a crucial role in the perception subsystem of autonomous driving systems (ADSs). However, studies reveal that the TSC system can make dangerous and potentially fatal errors under adversarial attacks. Existing defense strategies, such as adversarial training (AT), have demonstrated effectiveness but struggle to generalize across diverse attack scenarios. Recent advancements in self-supervised learning (SSL), particularly adversarial contrastive learning (ACL) methods, have demonstrated strong potential in enhancing robustness and generalization compared to AT. However, conventional ACL methods lack mechanisms to ensure effective defense transferability across different learning stages. To address this, we propose a robust pre-training–fine-tuning algorithm for meta-adversarial defense (RPF-MAD), designed to enhance the sustainability of adversarial robustness throughout the learning pipeline. Dual-track meta-adversarial pre-training (Dual-MAP) integrates meta-learning with ACL methods, which improves the generalization ability of the upstream model to different adversarial conditions. Meanwhile, adaptive variance anchoring robust fine-tuning (AVA-RFT) utilizes adaptive prototype variance regularization to stabilize feature representations and reinforce the generalizable defense capabilities of the downstream model. Leveraging the meta-adversarial defense benchmark (MAD) dataset, RPF-MAD ensures comprehensive robustness against multiple attack types. Extensive experiments across eight ACL methods and three robust fine-tuning (RFT) techniques demonstrate that RPF-MAD significantly improves both standard accuracy ($SA$) by 1.53% and robust accuracy ($RA$) by 2.64%, effectively enhances the lifelong adversarial resilience of TSC models, achieves a 13.77% improvement in the equilibrium defense success rate ($EDSR$), and reduces the attack success rate ($ASR$) by 9.74%, outperforming state-of-the-art (SOTA) defense methods. Full article

DNS over HTTPS (DoH) is an advanced version of the traditional DNS protocol that prevents eavesdropping and man-in-the-middle attacks by encrypting queries and responses. However, it introduces new challenges such as encrypted traffic communication, masking malicious activity, tunneling attacks, and complicating intrusion detection system (IDS) packet inspection. In contrast, unencrypted packets in the traditional Non-DoH version remain vulnerable to eavesdropping, privacy breaches, and spoofing. To address these challenges, an optimized dual-path feature selection approach is designed to select the most efficient packet features for binary class (DoH-Normal, DoH-Malicious) and multiclass (Non-DoH, DoH-Normal, DoH-Malicious) classification. Ant Colony Optimization (ACO) is integrated with machine learning algorithms such as XGBoost, K-Nearest Neighbors (KNN), Random Forest (RF), and Convolutional Neural Networks (CNNs) using CIRA-CIC-DoHBrw-2020 as the benchmark dataset. Experimental results show that the proposed model selects the most effective features for both scenarios, achieving the highest detection and outperforming previous studies in IDS. The highest accuracy obtained for binary and multiclass classifications was 0.9999 and 0.9955, respectively. The optimized feature set contributed significantly to reducing computational costs and processing time across all utilized classifiers. The results provide a robust, fast, and accurate solution to challenges associated with encrypted DNS packets. Full article

Conventional multi-object tracking approaches frequently exhibit performance degradation in marine radar (MR) imagery due to complex environmental challenges. To overcome these limitations, this paper proposes ShipMOT, an innovative multi-object tracking framework specifically engineered for robust maritime target tracking. The novel architecture features three principal innovations: (1) A dedicated CNN-based ship detector optimized for radar imaging characteristics; (2) A novel Nonlinear State Augmentation (NSA) filter that mathematically models ship motion patterns through nonlinear state space augmentation, achieving a 41.2% increase in trajectory prediction accuracy compared to conventional linear models; (3) A dual-criteria Bounding Box Similarity Index (BBSI) that integrates geometric shape correlation and centroid alignment metrics, demonstrating a 26.7% improvement in tracking stability under congested scenarios. For a comprehensive evaluation, a specialized benchmark dataset (Radar-Track) is constructed, containing 4816 annotated radar images with scenario diversity metrics, including non-uniform motion patterns (12.7% of total instances), high-density clusters (>15 objects/frame), and multi-node trajectory intersections. Experimental results demonstrate ShipMOT’s superior performance with state-of-the-art metrics of 79.01% HOTA and 88.58% MOTA, while maintaining real-time processing at 32.36 fps. Comparative analyses reveal significant advantages: 34.1% fewer ID switches than IoU-based methods and 28.9% lower positional drift compared to Kalman filter implementations. These advancements establish ShipMOT as a transformative solution for intelligent maritime surveillance systems, with demonstrated potential in ship traffic management and collision avoidance systems. Full article

With the rapid growth of heterogeneous fifth-generation (5G) communication networks and a surge in global mobile traffic, energy consumption in mobile network systems has increased significantly. This underscores the need for energy-efficient networks that lower operational costs and carbon emissions, leading to a focus on microgrids powered by renewable energy. However, accurately predicting base station traffic demand and optimizing energy consumption while maximizing green energy usage—especially concerning quality of service (QoS) for users—remains a challenge. This paper proposes a cooperative sleep and energy-sharing strategy for heterogeneous 5G base station microgrid (BSMG) systems, utilizing deep learning and an improved multi-objective evolutionary algorithm based on decomposition (MOEA/D). We present a reference scenario for a 5G BSMG system comprising a central and sub-base station microgrid. A prediction model was developed, integrating a convolutional neural network with a dual attention mechanism and bidirectional long short-term memory to determine the operational status of BSMGs. Our cooperative strategy addresses QoS requirements and uses the enhanced MOEA/D to improve performance. Numerical results indicate that our approach achieves significant energy savings while ensuring accurate predictions of BSMG energy demands through a multi-objective evolutionary algorithm based on decomposition. Full article

The present study aims to investigate the benefits of eco-driving in urban areas and on highways through an experiment conducted in a driving simulator. Within a group of 39 participants aged 18–30, multiple driving scenarios were conducted, both without and with eco-driving guides, to assess the impact of eco-driving behavior on environmental sustainability and safety outcomes. Data on pollutant emissions, including carbon dioxide (CO₂), carbon monoxide (CO), and nitrogen oxides (NO_x), as well as crash probability, were collected during the experiment. The relationships between driving behavior and pollutant emissions were estimated using linear regression models, while binary logistic regression models were employed to assess crash probability. The analysis revealed that eco-driving led to a significant reduction in pollutant emissions, with CO₂ emissions decreasing by 1.42%, CO by 98.2%, and NO_x by 20.7% across both urban and highway environments, with a more substantial impact in urban settings due to lower average speeds and smoother driving patterns. Furthermore, eco-driving reduced crash probability by 90.0%, with urban areas exhibiting an 86.8% higher crash likelihood compared to highways due to higher traffic density and more complex driving conditions. These findings highlight the dual benefit of eco-driving in reducing environmental impact and improving road safety. This study supports the integration of eco-driving techniques into transportation policies and driver education programs to foster sustainable and safer driving practices. Full article

The rapid evolution of intelligent vehicle technology has significantly advanced autonomous decision-making and driving safety. However, the challenge of predicting long-term trajectories in complex urban traffic persists, as traditional methodologies usually handle spatiotemporal attention mechanisms in isolation and are typically limited to short-term trajectory predictions. This paper proposes a Habit-based Temporal–Spatial Attention Long Short-Term Memory (HTSA-LSTM) network, a novel framework that integrates a dual spatiotemporal attention mechanism to capture dynamic dependencies across time and space, coupled with a driving style analysis module. The driving style analysis module employs Sparse Inverse Covariance Clustering and Spectral Clustering (SICC-SC) to extract driving primitives and cluster trajectory data, thereby revealing diverse driving behavior patterns without relying on predefined labels. By segmenting real-world driving data into fundamental behavioral units that reflect individual driving preferences, this approach enhances the model’s adaptability. These behavioral units, in conjunction with the spatiotemporal attention outputs, serve as inputs to the model, ultimately improving prediction accuracy and robustness in multi-vehicle scenarios. The model was evaluated by using the NGSIM dataset and real driving data from Wuhan, China. In comparison to benchmark models, HTSA-LSTM achieved a 20.72% reduction in the root mean square error (RMSE) and a 24.98% reduction in the negative log likelihood (NLL) for 5 s predictions of long-term trajectories. Furthermore, HTSA-LSTM achieved R² values exceeding 97.9% for 5 s predictions on highways and expressways and over 92.7% for 3 s predictions on urban roads, highlighting its excellent performance in long-term trajectory prediction and adaptability across diverse driving conditions. Full article

Pavement design is influenced by traffic load, which affects its lifespan. Traditional methods classify traffic load into fixed traffic, fixed vehicle, variable traffic, and vehicle/axle loads. In fixed traffic, pavement thickness is based on the maximum expected wheel load without considering load repetitions. Meanwhile, in fixed vehicle scenarios, it is calculated by the repetitions of a standard axle load. For nonstandard axle loads, the equivalent axle load is determined by multiplying repetitions by the corresponding equivalent load factor. In variable traffic, each axle and its repetitions are analyzed independently. This study proposes enhanced models for fixed traffic loads, focusing on single, dual, and tridem axles in a single-layer pavement model, to improve stress prediction accuracy. The results show that a quadratic model with a base-10 logarithmic transformation accurately predicts stresses. Additionally, machine learning models, especially Gradient Boosting, provided more accurate predictions than traditional models, with lower mean squared error (MSE) and root mean squared error (RMSE). The results show that these models are effective in predicting the stress in pavement. These findings provide valuable insights that can lead to better pavement design and more effective maintenance practices. Full article