Mobile Edge Computing (MEC) is a key technology for enabling energy-efficient and low-latency processing of computation-intensive applications through task offloading. However, current frameworks typically model dependent tasks using static Directed Acyclic Graphs (DAGs), which are poorly suited to dynamic edge environments characterized by fluctuating resources and diverse QoS demands. These fixed DAG structures often fail to adapt to runtime changes in bandwidth or node workload, leading to frequent task failures or inefficient resource usage. To overcome these limitations, we propose Dynamic Application Configuration, a paradigm that allows applications to switch between multiple predefined DAG variants at runtime. This enables the system to dynamically balance accuracy and resource efficiency at critical decision points by adapting to current network and computing conditions. Based on this concept, we design the Dynamic Application Configuration and Capacityaware task offloading System (DACCS), which employs a two-step offloading strategy: (i) a Dynamic Graph Selection (DGS) algorithm that adaptively adjusts application configurations during key subtask execution based on real-time resource states, and (ii) a dependency-aware offloading algorithm (DAP) that optimizes offloading assignments by jointly optimizing the application completion rate, processing delay and effective utilization rate. Experiments and real-system validations demonstrate that DGS improves the service performance across multiple strategies. Furthermore, the proposed DGS-DAP strategy outperforms other benchmark approaches under dynamic conditions.

As a prevalent field of study in machine learning, intelligent agents can perceive surroundings and make informed decisions. In many research areas such as autopilot systems, undersea explorations, and distributed robotics, researchers have traditionally employed unidirectional systems, which usually rely on perceptions from sensors to controllers, or end-to-end models, which generate actions directly from raw data. Nonetheless, in unidirectional systems, controller efficiency is intrinsically linked to sensor accuracy, which makes unidirectional systems lack a self-improving capability. Meanwhile, compared with functionally separated frameworks, end-to-end methods may have their own limitations in scalability, generality, interoperability, training costs, and so on. To fulfill this gap, we propose a new Coordinated Optimization Framework for Intelligent Agents (COIA). We introduce an inverted optimization channel from controllers to sensors in traditional functionally separated frameworks through communication between devices, enabling closed-loop online evolutive learning. To the best of our knowledge, this paper first presents a universal coordinated optimization framework among supervised learning and RL models, without human labels or intervention. Our method allows heterogeneous agents to autonomously adapt to some special situations in open environments, which forms a basis of networked Artificial General Intelligence (AGI). We design an experimental paradigm of COIA with concrete cases, which shows a significantly large performance margin over unidirectional and end-to-end models. The performance margin grows with task complexity.

The criticality of efficient traffic forecasting in Intelligent Transportation System (ITS) has garnered significant academic attention. This study addresses the prevalent issue of distribution shift in real-world datasets, which often degrades performance, and explores the effectiveness of the channel-independence (CI), a technique recently proposed to mitigate this issue. While Spatio-Temporal Graph Neural Networks (STGNNs) are noted for their flexibility to represent road structures, their designs typically lack the capability to integrate CI without disrupting the spatial relationships, potentially limiting the performance. We present a novel approach that successfully integrates CI into spatial-temporal forecasting by incorporating distinct temporal, spatial, and predefined graph structure information within each channel. Moreover, STGNNs frequently emphasize intricate designs, which result in increased computational demands while offering only marginal improvements in accuracy. This paper presents ST-MLP, a streamlined spatio-temporal model constructed exclusively from cascaded Multi-Layer Perceptron (MLP) modules and linear layers. Experimental results indicate that ST-MLP outperforms numerous existing STGNNs in both accuracy and computational efficiency. Our findings advocate for further investigation into more streamlined and effective neural network architectures within spatial-temporal forecasting research.

Traditional recurrent neural network architectures, such as long short-term memory neural networks (LSTM), have historically held a prominent role in time series forecasting (TSF) tasks. While the recently introduced sLSTM for Natural Language Processing (NLP) introduces exponential gating and memory mixing that are beneficial for long term sequential learning, its potential short memory issue is a barrier to applying sLSTM directly in TSF. To address this, we propose a simple yet efficient algorithm named P-sLSTM, which is built upon sLSTM by incorporating patching and channel independence. These modifications substantially enhance sLSTM's performance in TSF, achieving state-of-the-art results. Furthermore, we provide theoretical justifications for our design, and conduct extensive comparative and analytical experiments to fully validate the efficiency and superior performance of our model.

As a potential application of Vehicle-to-Everything (V2X) communication, multi-agent collaborative perception has achieved significant success in 3D object detection. While these methods have demonstrated impressive results on standard benchmarks, the robustness of such approaches in the face of complex real-world environments requires additional verification. To bridge this gap, we introduce the first comprehensive benchmark designed to evaluate the robustness of collaborative perception methods in the presence of natural corruptions typical of real-world environments. Furthermore, we propose DSRC, a robustness-enhanced collaborative perception method aiming to learn Density-insensitive and Semantic-aware collaborative Representation against Corruptions. DSRC consists of two key designs: i) a semantic-guided sparse-to-dense distillation framework, which constructs multi-view dense objects painted by ground truth bounding boxes to effectively learn density-insensitive and semantic-aware collaborative representation; ii) a feature-to-point cloud reconstruction approach to better fuse critical collaborative representation across agents. To thoroughly evaluate DSRC, we conduct extensive experiments on real-world and simulated datasets. The results demonstrate that our method outperforms state-of-the-art collaborative perception methods in both clean and corrupted conditions.

Real-time animal detection in search-and-rescue (SAR) operations represents a critical challenge for situational awareness systems, particularly when deploying lightweight solutions on resource-constrained edge computing platforms. Current detection methods suffer from computational bottlenecks that compromise either accuracy or real-time performance, limiting their effectiveness in time-critical rescue scenarios. This paper presents SVR-YOLO11, an enhanced detection network specifically optimized for real-time animal identification in SAR applications. The proposed architecture introduces three key in-novations: (1) a Slim-Neck design paradigm that preserves inter-channel connections while reducing computational complexity, (2) integration of Large Separable Kernel Attention (LSKA) modules with dynamic upsampling to improve detection accuracy for morphologically similar animal species across diverse natural backgrounds, and (3) direct deployment capability on edge devices without requiring complex model conversion processes. Experimental validation on the Animals. v4i dataset demonstrates that SVR-YOLO11 achieves superior performance with 68.3% mAP50-95 while maintaining only 2.843M parameters and 6.5 GFLOPs computational cost. Real-time testing on multiple edge computing platforms, including Raspberry Pi and NVIDIA Jetson devices, confirms the network's practical applicability for field deployment. The system's effectiveness is further validated through a comprehensive head-mounted display demonstration using Unity Engine simulations that replicate dynamic SAR environments. These results establish SVR-YOLO11 as a ro-bust solution for real-time animal detection in distributed edge computing scenarios, directly supporting enhanced situational awareness in critical rescue operations.

In this study, to reduce the harm of the social and economic impacts of typhoons, a novel CNN-Transformer En-coder hybrid model is proposed to predict the typhoon trajectory for the previous preparation of typhoons, designed to support real-time simulation systems and improve preparedness against typhoon impacts. It combines the spatial feature extraction ability of a Convolutional Neural Network (CNN) with the temporal dependency modeling ability of a Transformer Encoder. Two different datasets, including the CMA tropical cyclone best track dataset and the ERA5 dataset, are integrated to improve the prediction accuracy and the diversity of data sources. Further-more, by incorporating critical environmental factors such as sea pressure and sea surface temperature, the predictive model can be integrated into real-time typhoon simulation systems to en-hance their dynamic accuracy and operational applicability. The model is validated by performance metrics such as mean square error, root mean squared error, coefficient of determination and case studies, demonstrating its adaptability to incomplete data and its ability to predict accurate trajectories. Subsequently, a comparative analysis between the CNN- Transformer model and the CNN-LSTM model demonstrates the better performance and enhanced predictive capabilities of the proposed approach.

Accurate traffic flow prediction is crucial for intelligent transportation systems supporting emerging applications such as autonomous driving and vehicle-infrastructure cooperation. However, existing methods often struggle to effectively disentangle the inherent trend, seasonal, and noise components within traffic flow data, thereby limiting prediction accuracy. To address this issue, we propose MF-AttnBiLSTM, a novel hybrid framework combining signal processing and temporal attention-based deep learning model through a decompose-then-predict strategy. Our approach first employs moving average to extract the trend component and discrete Fourier transform to isolate dominant seasonal patterns from the residuals. Subsequently, a dual-stream architecture utilizes multi-head self-attention-enhanced bidirectional LSTMs to independently model the temporal dynamics of the decomposed trend and seasonal components. The final prediction aggregates the outputs from both streams. Extensive experiments on PeMS04 and PeMS07 datasets demonstrate that MF-AttnBiLSTM significantly outperforms state-of-the-art baselines and exhibits robustness across varying traffic conditions. Ablation studies further confirm the efficacy of each component, particularly highlighting the significant contribution of the signal decomposition stage to overall performance improvement.

Multi-agent collaborative perception enables a single agent to perceive the comprehensive scene by exchanging sensory information using vehicle-to-everything communication. However, the communication resources of real-world communication systems are often limited. It fails to satisfy the real-time transmission of extensive data, which restricts the deployment of collaborative perception. To address the issue, we propose ComSH, a Communication-efficient multi-agent collaborative perception method based on Spatio-temporal Heterogeneity to achieve a trade-off between performance and bandwidth. Specifically, we consider the spatio-temporal heterogeneity to perform feature filtering, thus reducing the redundancy of transmitted features. First, it introduces valuable temporal semantics to enhance the current representation of each agent. Secondly, we consider the spatial confidence of each agent at each location and the relative position to obtain high confidence and complementary features. Eventually, a foreground object adaptive activation module is designed to enrich the visual representation of fused features. Therefore, ComSH enables different agents to transmit their critical and complementary perception information, thus reducing bandwidth consumption. We conduct experiments on collaborative detection tasks on the two datasets. Experimental results demonstrate the ComSH’s superiority and effectiveness.

Crime prediction (CP) plays a pivotal role in urban analytics, contributing significantly to personal safety and societal stability. Unlike conventional time series forecasting, CP faces unique difficulties due to the inherent sparsity of crime incidents, particularly within small spatial regions and limited time windows. This sparsity, coupled with the non-Gaussian distribution of crime data—characterized by an excess of zero events and over-dispersion—presents a critical challenge for the signal processing community. In this regard, we propose a novel framework, Spatial-Temporal Multivariate Zero-Inflated Negative Binomial Graph Neural Networks (STMGNN-ZINB), which integrates diffusion and convolutional graph networks to capture spatial, temporal, and multivariate dependencies. By leveraging a Zero-Inflated Negative Binomial distribution, the STMGNN-ZINB effectively models the over-dispersed and zero-heavy nature of crime data, significantly improving both prediction accuracy and confidence interval estimation. Experimental results on real-world datasets demonstrate that our STMGNN-ZINB outperforms state-of-the-art CP methods, offering a robust tool for crime early warning and explicable insights into urban crime dynamics.

Autonomous driving (AD) systems requires multisensor collaboration to address challenges caused by complex weather scenarios. The recognition accuracy of autonomous driving system based on single resource, either on image only or Lidar only, becomes unreliable due to untested weather conditions, occlusions objects, and other factors. This paper proposes a decision-level fusion network based on an improved YOLOV7 and an improved CenterPoint network to build a multi-sensor-collaboration scheme. The overall accuracy of the proposed fusion algorithm is improved for small targets by adding multi-scale and multi-stage attention channel modules into the backbones of image recognition network and point cloud recognition network respectively. Moreover, the fusion algorithm introduces mixed distance constraints as the loss function for overlapping targets. The proposed fusion algorithm has been successfully tested on the public ONCE dataset mixed with a self-built dataset under various road conditions such as sunny, night, and rainy weather. The mAP of proposed decision-level fusion algorithm achieves$\mathbf{8 3. 5 \%}$in sunny daytime,$\mathbf{8 0 \%}$during nighttime and 79.1 % during rain time.

In the contemporary digital landscape, individuals are often overwhelmed by the plethora of choices available, including articles, the focus of this article. This paper explores recommender systems aimed at enhancing diversity and mitigating the homogeneity often found in traditional systems to help reduce the risk of filter bubbles and stereotypes, ensuring fair and inclusive user experiences and with little cold start problem as an item-based system. Compared to a Maximal Marginal Relevance (MMR) baseline, the GA-based system is adept at providing relevant and diverse recommendations, particularly when generating a large set of options based on limited input. In addition, the operational speed of the genetic algorithm (GA) due to its constant-time response to varying recommendation sizes underscores its practical advantages. For customizability, the system’s linear and predictable response to adjustments in the trade-off parameters enhances its adaptability to different user needs. These advancements make the GA-based recommender system a potent tool for addressing the complexity and diversity of user preferences in large-scale applications, presenting a significant step forward in the development of recommender systems.

In artificial intelligence, enhancing agents’ perception and decision-making in physical environments is a key research focus. Perception and control, as essential components of task execution, interact closely and jointly determine system performance. Traditionally, agents use perception models to extract environmental information and control algorithms to generate actions. However, in real-world scenarios, the choice and combination of perception and control models can greatly affect efficiency and accuracy, demanding stronger adaptability and robustness. To study these effects, we design a simulation platform for path planning tasks built on the OpenAI Gym framework. The platform is flexible and scalable, supporting integration and evaluation of diverse perception-control algorithm combinations. Based on this platform, we conduct experiments with different perception-control pairings and further propose a perception optimization mechanism driven by control feedback, namely online evolutive learning mechanism. This mechanism dynamically adjusts the perception model within the closed-loop system, enabling agents to adapt online and respond more effectively to environmental changes. Experimental results show that perception-control combinations yield notable performance differences under identical conditions. Moreover, the proposed optimization mechanism significantly accelerates convergence and improves execution efficiency in non-stationary environments, confirming its effectiveness in enhancing system robustness and adaptability.

Video anomaly detection (VAD) is an essential task in the image processing community with prospects in video surveillance, which faces fundamental challenges in balancing detection accuracy with computational efficiency. As video content becomes increasingly complex with diverse behavioral patterns and contextual scenarios, traditional VAD approaches struggle to provide robust assessment for modern surveillance systems. Existing methods either lack comprehensive spatial-temporal modeling or require excessive computational resources for real-time applications. In this regard, we present a Mamba-based multi-scale spatial-temporal learning (M2S2L) framework in this paper. The proposed method employs hierarchical spatial encoders operating at multiple granularities and multi-temporal encoders capturing motion dynamics across different time scales. We also introduce a feature decomposition mechanism to enable task-specific optimization for appearance and motion reconstruction, facilitating more nuanced behavioral modeling and quality-aware anomaly assessment. Experiments on three benchmark datasets demonstrate that M2S2L framework achieves 98.5%, 92.1%, and 77.9% frame-level AUCs on UCSD Ped2, CUHK Avenue, and ShanghaiTech respectively, while maintaining efficiency with 20.1G FLOPs and 45 FPS inference speed, making it suitable for practical surveillance deployment.

Enabling Mobile edge computing (MEC) services with massive connectivity and low energy consumption is crucial for future Internet of Things (IoT) infrastructures. In this article, we investigate an IoT network, where MEC is adopted as the computing framework for complicated IoT services while nonorthogonal multiple access (NOMA) is introduced to enable the interdependent data input offloading from multiple IoT devices to an edge server. The MEC service frame consists of a communication phase and a computation phase, where the latter phase requires the complete data offloaded in the former one to complete a specific task. To minimize the weighted sum energy consumption of both users and the edge server, a joint resource allocation original problem is formulated, which is unfortunately nonconvex. To tackle the difficulty, we first decompose the problem into subproblems, and characterize the structure of optimal solution to the subproblems. Following the characterization, the original problem is reformulated into a tractable one. We then develop a Branch-Reduce-and-Bound (BRB) based algorithm to obtain the optimal solution. Additionally, to further investigate the MEC scenario with the offloading of multiple users, we apply the state-of-the-art hybrid NOMA (H-NOMA) scheme to evaluate its benefits to multiuser MEC. We rigorously prove that, with interdependent user data inputs, pure NOMA (P-NOMA) is not only a special case but also an optimal case of H-NOMA. Via simulation, the analytical findings and the proposed algorithm are validated and evaluated.

Video Anomaly Detection (VAD) remains a fundamental yet formidable task in the video understanding community, with promising applications in areas such as information forensics and public safety protection. Due to the rarity and diversity of anomalies, existing methods only use easily collected regular events to model the inherent normality of normal spatial-temporal patterns in an unsupervised manner. Although such methods have made significant progress benefiting from the development of deep learning, they attempt to model the statistical dependency between observable videos and semantic labels, which is a crude description of normality and lacks a systematic exploration of its underlying causal relationships. Previous studies have shown that existing unsupervised VAD models are incapable of label-independent data offsets (e.g., scene changes) in real-world scenarios and may fail to respond to light anomalies due to the overgeneralization of deep neural networks. Inspired by causality learning, we argue that there exist causal factors that can adequately generalize the prototypical patterns of regular events and present significant deviations when anomalous instances occur. In this regard, we propose Causal Representation Consistency Learning (CRCL) to implicitly mine potential scene-robust causal variable in unsupervised video normality learning. Specifically, building on the structural causal models, we propose scene-debiasing learning and causality-inspired normality learning to strip away entangled scene bias in deep representations and learn causal video normality, respectively. Extensive experiments on benchmarks validate the superiority of our method over conventional deep representation learning. Moreover, ablation studies and extension validation show that the CRCL can cope with label-independent biases in multi-scene settings and maintain stable performance with only limited training data available.

Collaborative perception systems enhance the perception capabilities of individual vehicles by facilitating information exchange between neighbouring vehicles. This approach effectively addresses challenges like occlusions and long-range perceptions that single vehicle cannot manage alone. However, practical applications often face difficulties due to constraints in wireless communication resources and reliability, which limit the effectiveness of latency-sensitive collaborative perception. To overcome these barriers, we introduce CERCP, a Communication Efficient and Robust Collaborative Perception framework. CERCP comprises two core modules: a cross-vehicle spatio-temporal feature selection module, which minimizes communication by transmitting only essential sensor regions with spatio-temporal complementarity, and a global-aware feature synchronization module, which mitigates data delays due to communication latency. To our knowledge, CERCP is the first general collaborative perception framework designed for efficient communication and is applicable across various tasks and modalities. We comprehensively evaluate CERCP on three datasets from real-world and simulated scenarios, using two sensor modalities (LiDAR and camera) and two perception tasks (3D object detection and BEV semantic segmentation). Extensive experiments demonstrate the superior performance of our method.

Collaborative perception enhances perception performance by enabling autonomous vehicles to exchange complementary information. Despite its potential to revolutionize the mobile industry, challenges in various environments, such as communication bandwidth limitations, localization errors and information aggregation inefficiencies, hinder its implementation in practical applications. In this work, we propose ERMVP, a communication-Efficient and collaboration-Robust Multi-Vehicle Perception method in challenging environments. Specifically, ERMVP has three distinct strengths: i) It utilizes the hierarchical feature sampling strategy to abstract a representative set of feature vectors, using less communication overhead for efficient communication; ii) It employs the sparse consensus features to execute precise spatial location calibrations, effectively mitigating the implications of vehicle localization errors; iii) A pioneering feature fusion and interaction paradigm is introduced to integrate holistic spatial semantics among different vehicles and data sources. To thoroughly validate our method, we conduct extensive experiments on real-world and simulated datasets. The results demonstrate that the proposed ERMVP is significantly superior to the state-of-the-art collaborative perception methods.

Autonomous driving technology has become increasingly important in recent years, with the potential to revolutionize transportation systems and improve road safety. Vision-based methods have long been used in this field, but the major challenges in object detection are efficiency and occlusion. To address this challenge, anchor-free collaborative detection has been proposed as a promising solution. Despite its potential, there has been limited research on this approach. This study proposes an efficient vision-based multi-view object detection and localization method that leverages anchor-free collaborative detection to improve the accuracy of pedestrian detection. The method first generates feature maps to extract the head and foot of pedestrians and then applies spatial aggregation to fuse information from different views. Additionally, the study examines the efficiency of different convolutional neural network architectures for the feature map extraction model and identifies ResNet18 and ResNet34 as the most efficient models for the task. The proposed method has the potential to significantly improve the accuracy of pedestrian detection and localization in autonomous driving scenarios, which is critical for ensuring safety. Overall, this work contributes to the development of vision-based methods for autonomous driving and has significant implications for the future of transportation technology.

As one of the research objects in machine learning, agents are endowed with the ability to perceive and make decisions. In order to meet the needs of different applications, many researchers focus on algorithms for controlling agents. In most current paradigms, the control algorithms take the information of the surrounding environment as input and output actions. However, under the influence of this unidirectional data flow, the upper limit of the performance of control algorithms depends on the accuracy and fit measure of the environment information. Even though recent end-to-end control algorithms take the environmental raw data as input and reduce the influence of perception performance on it, the raw data acquisition method is still fixed, therefore, the performance of these control algorithms is still limited by the environmental information acquisition scheme. Meanwhile, although most control methods have the ability to update parameters online, they usually do not have the ability to optimize the acquisition of environmental information, and there may be unknown situations that are not in the data set used for pre-training. As a result, current methods cannot handle detection inaccuracies and unknown situations, and lack the ability to further improve their own performance online. In order to resolve the deficiency of the traditional control paradigms, we introduce the reverse optimization channel from the controller to the environmental information acquisition scheme to form a new algorithm through communication between devices. Experiments show that our method has significant performance margin and universal online evolutive learning ability, as compared to traditional paradigms.

This article investigates the influence of limited information on the efficacy of traffic signal control algorithms for supporting Intelligent Transportation Systems. Our project has collected several classic and trending intelligent traffic signal control schemes and evaluated algorithmic performance under constrained data conditions. In these simulating scenarios, access to certain traffic data is restricted. The findings reveal that although information scarcity generally degrades algorithm performance, algorithms that perform well with comprehensive data maintain their superiority even in data-limited environments. This underscores the necessity of designing resilient algorithms capable of adapting to varying levels of information availability, offering valuable insights for developing robust traffic control systems.

COVID-19 has become the most devastating infectious disease of the 21st century. Governments worldwide devised a range of policies to control the pandemic and reduce losses in multiple aspects. To make timely and wise decisions, a thorough analysis of policies with reliable statistical evidence is crucial. Obtaining these statistical references to policies would help decision-makers optimize intervention policies to the maximum. In this work, we designed a policy-aware time series forecasting model based on the transformer architecture and successfully estimated COVID-19 epidemic trends. Through incorporating temporal information from 16 policy indicators, we developed a policy-aware time series model that demonstrated high forecasting performance. We further quantify the causal effect of indicators by employing a counterfactual approach and propose two static metrics lag period and average effect. The empirical results demonstrate that our model causally verifies the effectiveness of all 16 policy indicators in controlling COVID-19 virus transmission in the US. From qualitative analysis, we conclude that frequent adjustments to the extent of policy actions may intensify epidemic spread. Our study provides insight into modeling government policies from a statistical angle, and it has practical applications when confronting potential influenza-like diseases in the future.

Intelligent Transportation Systems (ITS) have garnered considerable attention as a potential solution for addressing the conflict between the increasing demand for transportation and the constraints within transportation infrastructure. One pivotal facet of this field is the domain of traffic flow prediction. In this paper, we introduce an inventive methodology for traffic flow prediction, in which we employ CNN to capture the underlying traffic data trends, while Support Vector Regression (SVR) with the signature kernel is adapted to predict the residual components within the traffic data. We evaluated our approach through comprehensive experiments based on real world traffic data, and the results clearly demonstrate a significant improvement in prediction accuracy over both ablation models and alternative state-of-art baseline methods.

The ever-changing traffic dynamics make the traditional traffic signal control methods unable to adapt to the environment. Meanwhile, deep reinforcement learning (DRL) has the property of interacting with the environment and adapting to changes in the environment. Therefore, in recent years, researchers have usually solved traffic signal control (TSC) problems through DRL methods. They have not only improved the design of neural networks, but also improved the ability of models to understand traffic conditions and learn corresponding task requests by designing different states and rewards. However, although the existing TSC algorithms based on DRL have proposed many well-designed states and reward strategies, which combinations of states and rewards should be adopted in practice to achieve the performance margin of models remains a question that researchers are seeking the answer to. Therefore, we introduce a general simulation platform to test and compare experimental performance under different combinations of states and rewards. Specifically, we test and analyze the experimental effects under different combinations of multiple traffic states and rewards through various TSC methods with a set of unified model settings. We further design and test some new state representations and reward strategies based on more detailed traffic information. The test results show that when researchers design the state and reward, refining the traffic state like vehicle running condition and making the state and reward match can make the experimental performance better than other combinations in most cases. We hope these results have some implications for the state and reward choice when researchers conduct experiments on TSC problem or other traffic decision management problems.

As essential tools for industry safety protection, automatic video anomaly detection systems (AVADS) are designed to detect anomalous events of concern in surveillance videos. Existing VAD methods lack effective exploration of the prototypical appearance and motion features leading to poor performance in realistic scenarios. Specifically, they either misreport regular events as anomalies due to insufficient representation power, or lead to missed detections with over-power generalization. In this regard, we propose an appearance-motion prototype network (AMP-net) that uses external memories to record prototype features and augments the appearance-motion prototype with a spatial-temporal fusion. In addition, AMP-net sequentially fuses appearance features from deep to shallow to utilize multiscale spatial context. Additionally, we introduce temporal attention to capture important dynamics and enhance AMP-net for representing regular motion. The proposed method achieves a delicate balance of effective representation of normal events and limited generalization to anomalies. Experiments on three benchmark datasets demonstrate that our method can accurately detect anomalous events, achieving performance comparable to state-of-the-art methods with frame-level AUCs of 98.7%, 92.4%, and 78.8% on the UCSD Ped2, CUHK Avenue, and ShanghaiTech datasets. Moreover, we conducted a case study on the self-collected industrial dataset, and the results indicate that our AMP-net can cope with complex industrial scenarios and outperform existing methods.

With the rapid development of intelligent transportation system applications, a tremendous amount of multi-view video data has emerged to enhance vehicle perception. However, performing video analytics efficiently by exploiting the spatial-temporal redundancy from video data remains challenging. Accordingly, we propose a novel traffic-related framework named CEVAS to achieve efficient object detection using multi-view video data. Briefly, a fine-grained input filtering policy is introduced to produce a reasonable region of interest from the captured images. Also, we design a sharing object manager to manage the information of objects with spatial redundancy and share their results with other vehicles. We further derive a content-aware model selection policy to select detection methods adaptively. Experimental results show that our framework significantly reduces response latency while achieving the same detection accuracy as the state-of-the-art methods.

Vehicular Edge Computing (VEC), as one of the major components of Intelligent Transportation Systems, improves road safety by providing computing services to safety-related applications on vehicles. Currently, the existing fine-grained computing scheduling algorithms are normally designed based on some simple scheduling policies. Due to the heterogeneous nature of tasks offloaded from various applications, they may not effectively satisfy various performance requirements of the real system, thereby leading to the problem that the short-term residual computing power cannot be effectively utilized when computing-costly tasks occupy the server. Therefore, improving the overall system performance and the efficiency of utilizing computing power is a critical issue. Accordingly, in this paper, we study the problem of computing scheduling inside edge servers in VEC, where multiple tasks can be offloaded to Road Side Units (RSUs). We analyze the role played by multiple evaluation metrics in the existing methods for ensuring the quality of service (QoS) and further design a novel online multi-factor aware task offloading algorithm with a hierarchical fine-grained computing scheduling scheme inside the edge server. We evaluate it by conducting intensive simulation tests and comparing the results with some state-of-the-art approaches. Numerical results show that the proposed algorithm outperforms the methods in the control group in different aspects and achieves the best overall performance.

Multi-agent collaborative perception as a potential application for vehicle-to-everything communication could significantly improve the perception performance of autonomous vehicles over single-agent perception. However, several challenges remain in achieving pragmatic information sharing in this emerging research. In this paper, we propose SCOPE, a novel collaborative perception frame-work that aggregates the spatio-temporal awareness characteristics across on-road agents in an end-to-end manner. Specifically, SCOPE has three distinct strengths: i) it considers effective semantic cues of the temporal context to enhance current representations of the target agent; ii) it aggregates perceptually critical spatial information from heterogeneous agents and overcomes localization errors via multi-scale feature interactions; iii) it integrates multi-source representations of the target agent based on their complementary contributions by an adaptive fusion paradigm. To thoroughly evaluate SCOPE, we consider both real-world and simulated scenarios of collaborative 3D object detection tasks on three datasets. Extensive experiments show the superiority of our approach and the necessity of the proposed components. The project link is https://ydk122024.github.io/SCOPE/.

This work proposes a novel three-layer federated learning (FL) framework with parameter selection and pre-synchronization (PSPFL) to achieve fast and accurate model training. The basic idea of PSPFL is that clients select partial model parameters for transmission and then base stations aggregate them cooperatively (i.e., pre-synchronization) and send the aggregated results to the server for global model update periodically. However, there is an intrinsic trade-off between parameter transmission overhead and model training loss. To strike a desirable balance between them, we investigate the optimal parameter pre-synchronization round and local training round under PSPFL. Specifically, we propose a Deep Q-Network (DQN)-based method to obtain the local training round and parameter pre-synchronization round. Finally, extensive experiments are conducted to evaluate the performance of the proposed method on commonly used datasets. The results show that the proposed method can reduce the sum of FL completion time and training loss by an average of 8.17%-18.82% compared to benchmarks.

With the rise of advanced applications based on Artificial Intelligence (AI) and Internet-of-Things (IoT), mobile devices have become more intelligent, introducing a novel concept, Mobile Edge Intelligence. But the limited on-board resources often hinder the capabilities of mobile devices. Mobile Edge Computing (MEC), regarded as an effective method to expand device capability, effectively overcomes this barrier. However, the dynamic networks driven by mobility and the dependency on applications pose significant challenges for offloading, which can degrade MEC’s overall performance. Therefore, how to effectively combine the above points to achieve a stable and effective sharing of computing resources between devices and servers is a critical issue. In this paper, we consider a multi-slot MEC system with device mobility and multiple applications of unknown arrival. To improve application completion rate while reducing task delay, we introduce a novel, robust distributed offloading algorithm, which calls the Multi-Attention Pointer network-based Reinforcement Learning algorithm (MAPRL), for the dynamic and unstable resource offloading scenario. Numerous experiments have been carried out to demonstrate that, compared with the existing methods, MAPRL exhibits robustness when facing the changing scenario, it can adapt to the unknown workload and dynamic network connections to enhance the offloading performance.

Multi-agent collaborative perception has received increasing attention recently as an emerging application in driving scenarios. Despite advancements in previous approaches, challenges remain due to redundant communication patterns and vulnerable collaboration processes. To address these issues, we propose What2comm, an end-to-end collaborative perception framework to achieve a trade-off between perception performance and communication bandwidth. Our novelties lie in three aspects. First, we design an efficient communication mechanism based on feature decoupling to transmit exclusive and common feature maps among heterogeneous agents to provide perceptually holistic messages. Secondly, a spatio-temporal collaboration module is introduced to integrate complementary information from collaborators and temporal ego cues, leading to a robust collaboration procedure against transmission delay and localization errors. Ultimately, we propose a common-aware fusion strategy to refine final representations with informative common features. Comprehensive experiments in real-world and simulated scenarios demonstrate the effectiveness of What2comm.

The Traveling Salesman Problem (TSP) is an important NP-Hard combinatorial problem worth studying in Computer Science, Mathematical Optimization, and Operations Research. Heuristic methods are often employed in looking for a nearly-optimized solution for TSP. Common heuristic methods for solving the TSP are the Genetic Algorithm (GA), Ant Colony Optimization (ACO), Simulated Annealing (SA), and the Lin-Kernighan-Helsgaun (LKH) algorithm. Recent studies have combined the Reinforcement Learning (RL) methods with LKH to boost performance. However, little is known about the performance differences between the many heuristic methods given. Also, little is understood about the impact of the TSP landscape on the performance of the heuristic methods. In this paper, five heuristic methods of GA, ACO, SA, LKH, VSR-LKH are discussed. We employed four types of landscape instances from the TSPLIB to perform an empirical analysis of the five heuristics. For the TSP landscape instances tested, we find that the loss and convergence speed of the algorithms is not directly related to the dimension of the TSP instances but the landscape of the TSP instances instead.

As a potential solution to the growing conflict between the increasing demand for transportation and the limited capacity of transportation infrastructure, Intelligent Transportation Systems have gained considerable attention for their effectiveness in improving the efficiency of existing transportation infrastructure and enhancing traffic safety. Among various research areas, traffic flow prediction is a vital application, and researchers have devoted a lot of effort to designing accurate and fast algorithms. Currently, to satisfy various performance requirements, hybrid prediction methods that can take advantage of different sub-modules are beginning to emerge and show advantages in prediction accuracy and timeliness over other prediction algorithms that rely solely on machine learning. In this paper, we introduce a novel high precise traffic flow prediction method by utilizing the Fourier analysis (FA)-assisted denoising. Briefly, three sub-modules are introduced. Singular Spectrum Analysis (SSA) module is able to filter the noise of the original data, FA module is applied to extract periodic features of the traffic flow, and Long Short-Term Neural Networks (LSTM) is utilized to predict the future trend of time series residuals. We conducted simulation experiments. The corresponding test results demonstrate a substantial improvement in the accuracy compared to pure sub-models and other machine learning methods.

In recent years, people have paid more and more attention to the operational efficiency of the transportation system and the corresponding traffic safety and energy consumption issues. As a possible solution, with the improvement of vehicle detection capabilities and the corresponding computing power of vehicle computing equipment, the development of autonomous driving technology as an important application of AI technology in the automotive industry has attracted considerable attention from academics and industry. Technically, we consider vehicles and pedestrians to be the two most important participants of the transportation system. Accordingly, efficient coordination of relative movement between on-road vehicles and pedestrians plays a critical role in improving road safety. The effective detection of pedestrians from the vehicles’ point of view is fundamental for achieving such coordination. Therefore, in this paper, we will first provide a comprehensive study of existing pedestrian detection methods, especially the occluded pedestrian detection method. Then, we will discuss the potential solution for designing a practical pedestrian detection method for supporting autonomous driving.

How to effectively improve the traffic efficiency of the road network plays a crucial role in ensuring the regular operation of modern society. This is also a key concern in the field of intelligent transportation systems. As the basis for formulating traffic control strategies, efficient and accurate traffic flow forecasting is essential. Accordingly, various prediction methods have been proposed for addressing the traffic flow prediction issue. However, we notice that most researchers only take the accuracy performance as the primary evaluation criteria and do not consider the problem of time cost. Consequently, the timeliness of the prediction results cannot be guaranteed. In this case, no matter how high the accuracy of the prediction is, it cannot provide practical information for the formulation of traffic measures. Therefore, in this paper, by exploiting the dimension reduction ability of Auto-Encoder (AE), we proposed a time-efficient prediction method for a large-scale road network that significantly reduces the prediction processing time while ensuring prediction accuracy. We conducted simulation experiments, and the corresponding test results demonstrate a substantial improvement in the time efficiency of our method compared to the traditional methods.

In recent years, data-driven intelligent transportation systems (ITS) have developed rapidly and brought various AI-assisted applications to improve traffic efficiency. However, these applications are constrained by their inherent high computing demand and the limitation of vehicular computing power. Vehicular edge computing (VEC) has shown great potential to support these applications by providing computing and storage capacity in close proximity. For facing the heterogeneous nature of in-vehicle applications and the highly dynamic network topology in the Internet-of-Vehicle (IoV) environment, how to achieve efficient scheduling of computational tasks is a critical problem. Accordingly, we design a two-layer distributed online task scheduling framework to maximize the task acceptance ratio (TAR) under various QoS requirements when facing unbalanced task distribution. Briefly, we implement the computation offloading and transmission scheduling policies for the vehicles to optimize the onboard computational task scheduling. Meanwhile, in the edge computing layer, a new distributed task dispatching policy is developed to maximize the utilization of system computing power and minimize the data transmission delay caused by vehicle motion. Through single-vehicle and multi-vehicle simulations, we evaluate the performance of our framework, and the experimental results show that our method outperforms the state-of-the-art algorithms. Moreover, we conduct ablation experiments to validate the effectiveness of our core algorithms.

Creating a complete stylized font library that helps the audience to perceive information from the text often requires years of study and proficiency in the use of many professional tools. Accordingly, automatic stylized font generation in a deep learning-based fashion is a desirable but challenging task that has attracted a lot of attention in recent years. This paper revisits the state-of-the-art methods for stylized font generation and presents a taxonomy of the deep learning-based stylized font generation. Despite the notable performance of the existing models, stylized multilingual font generation, the task of applying specific font style to diverse characters in multiple languages has never been reported to be addressed. An efficient and economical method for stylized multilingual font generation is essential in numerous application scenarios that require communication with international audiences. We propose a solution for few-shot multilingual stylized font generation by a fast feed-forward network, Multilingual Font Generation Network (MF-Net), which can transfer previously unseen font styles from a few samples to characters from previously unseen languages. Following the Generative Adversarial Network (GAN) framework, MF-Net adopts two separate encoders in the generator to decouple a font image's content and style information. We adopt an attention module in the style encoder to extract both shallow and deep style features. Moreover, we also design a novel language complexity-aware skip connection to adaptive adjust the structural information to be preserved. With an effective loss function to improve the visual quality of the generated font images, we show the effectiveness of the proposed MF-Net based on quantitative and subjective visual evaluation, and compare it with the existing models in the scenario of stylized multilingual font generation. The source code is available on https://github.com/iamyufan/MF-Net.

We consider an update-and-decide IoT-based wireless network, where information packets generated from dual sources are co-stored in the transmitter's buffer, while decisions are made at the destination. Two practical assumptions about the communications between the transmitter and destination are taken into account: the communications are operating with finite blocklength (FBL) codes, and truncated hybrid automatic repeat request (HARQ) schemes are exploited to improve the FBL reliability, i.e., the number of allowed rounds of (re)transmissions is finite. For the first time, this paper characterizes the timeliness of status updates, namely age upon decisions (AuD) (which highlights the timeliness of the information at decisions in comparison to the concept of age of information), for such truncated HARQ-assisted wireless network. First, we characterize the inter-arrival time between two adjacent successfully transmitted packets, while taking into consideration the preemption policy and the randomness of the number of preempted packets from the same source. In particular, the probability density function, statistical performance of such inter-arrival time are derived. Following these characterizations, we propose a new approach to determine the average AuD and obtain a closed-form expression accordingly. Via simulations, we evaluate the performance and conclude a set of guidelines for designs on the considered network.

How to effectively improve the efficiency of vehicle traffic in the road system will play an essential role in improving the operational efficiency of the traffic system while eliminating the energy consumption and environmental pollution problems caused in particular, and this is also a key concern in the field of intelligent transportation systems. Timely and accurate traffic flow prediction is regarded as the key to solve the above problems because it can effectively improve the efficiency of traffic flow management. Many prediction methods have been proposed and among them, Machine Learning (ML)-based forecasting methods have gradually become mainstream in recent years because of their inherent ability to learn and predict nonlinear features in traffic information. However, we notice that most of the existing ML-based traffic prediction methods were designed relying fully on historical data while ignoring the structure and the impacts of the whole road network. Therefore, in this paper, we proposed a mixed method to take both historical data and road networks into consideration. Based on the real-world dataset, we conducted simulation experiments. The corresponding test results demonstrate a substantial improvement in the prediction accuracy of our method compared to conventional ML-based methods.

For developing a sustainable wireless sensor network (WSN), wireless energy charging and mobile data collection are two promising techniques for enhancing energy efficiency and achieving semi-permanent operation time of WSNs. Currently, many energy-aware methods have been developed by adopting these two technologies. However, joint methods that can combine the advantages of both are still lacking. Technically, by reducing the transmission energy consumption of individual nodes while wirelessly charging low power nodes, the joint method can better extend the operating time of the nodes in the system, improving the overall system life. Therefore, we design a two-mode QoS-aware mobile chargers (MCs) scheduling method for implementing both energy charging and data collection tasks simultaneously. The proposed scheme is comprised of two parts: 1) A new clustering algorithm is designed for addressing the trade-off problem regarding the delay and load balancing of sensors, and the network topology construction. 2) Two heuristic MC scheduling algorithms are introduced for facing the different delay requirements of systems, i.e., the single-path scheduling scheme (SPSS) and the multiple-path scheduling scheme (MPSS). The simulation results show that our proposed method outperforms the existing state-of-the-art approaches in the control group regarding the average delay and the charging utility.

Vehicle fuel efficiency (VFE) has a pivotal role in solving energy shortage issue due to the increasing global demand for energy. The high frequency of go-stop movements and long waiting times at intersections significantly reduce the VFE. Such negative impacts are particularly severe when the traffic flows are regulated by poorly designed traffic signal control. Existing works have successfully applied deep reinforcement learning (DRL) techniques to improve the efficiency of traffic signal control. However, to the best of our knowledge, few studies have explored traffic signal control for VFE through eco-driving techniques. To fill the gap, we propose a DRL-based fuel-economic traffic signal control for improving vehicle fuel efficiency. Briefly, we adopt the DRL-technique to develop an agent that can efficiently control traffic signals based on real-time traffic information at intersections, and adjust speed profiles for approaching vehicles to smooth traffic flows. We tested our method on both synthetic traffic dataset and real-world traffic dataset from surveillance cameras in Toronto. Through comprehensive experiments, we demonstrate that our method surpassed the performance of both pure eco-driving and pure traffic signal control techniques by significantly reducing vehicle fuel consumption and improving the efficiency of traffic signal control.

As the most serious global infectious disease in the past 100 years, it has caused severe loss of life and property to countries and their people worldwide in the past year. As the most powerful tool in the fight against the epidemic, how to quickly promote the COVID-19 vaccine administration plays a vital role in gradually establishing an immune barrier in the population as soon as possible and blocking the COVID-19 epidemic. In this paper, we provide a machine learning-based policy recommendation method on the vaccination campaign of COVID-19 by minimizing three different cost factors: the duration of the pandemic, the budget of the COVID-19 battle as well as the death toll. To generate a more efficient vaccination policy, we construct an Age-stratified Susceptible-Infected-Recovered (ASSIR) model. We validate our method based on the real-world dataset of India by comparing our simulated results with the government's vaccination plan from machine learning prediction. Our approach shows a 13% decrease in disease control time and government budget. At the same time, we find out that vaccination based on each province's population leads to a 12.4% decrease in the death toll than on infection cases. The model developed in this study has practical implications for COVID-19 vaccination campaigns and the infection control of other infectious diseases.

In recent years, with the continuously increasing number of vehicles, how to solve the frequent traffic accidents, the increasing traffic congestion, and the corresponding exhaust pollution in the transportation system is the problem that must be solved to ensure people's safe, efficient, and green travel needs. As one of the core components of future intelligent transportation systems (ITS), autonomous vehicles have become a common area of interest for academia and industry because they can strictly follow traffic laws and regulations while avoiding traffic accidents caused by improper driving behavior of human drivers. However, the current autonomous driving technology often relies on a single vehicle to independently detect its surrounding traffic environment. Under the impression of the detection range of the corresponding detector and the occlusion of different types of objects in the traffic system, a single vehicle often has a large detection blind spot. As a result, it may not be possible to develop an effective driving strategy in complex environments. For addressing this issue, in this article, we will introduce a collaborative object detection and warning method based on vehicular edge computing (VEC) to achieve a transparent traffic environment. In other words, by exploiting the computing power provided by the VEC, we will eliminate the detection blind spots of traffic system participants as much as possible. The efficiency of the proposed method will be evaluated through physical experiments.

In recent years, to improve the transportation system's efficiency, relying on the development of vehicular wireless communication technology and corresponding in-vehicle sensor technology, intelligent transportation systems have become the focus of attention in academia and industry. This is because, by improving the vehicle's ability to perceive the surrounding traffic environment through the exchange of information between vehicles, the vehicle's safety can be effectively improved, which reduces the occurrence of traffic accidents, in turn improving the efficiency of the transportation system. However, while data exchange brings convenience, similar to the other data communication applications, communication security issues inevitably enter the Internet-of-Vehicles environment since the vehicle is no longer an isolated individual. Accordingly, in this paper, we will propose a data transmission security improvement method based on accurate traffic flow prediction for addressing a specific data theft problem. Briefly, our method can detect the fraud vehicle that spread fake traffic information to increase the possibility that it may be selected as a relay node, thereby increasing its theft of the data of related users who use it as a relay node. Intensive simulation experiments are conducted to evaluate and prove the efficiency of our proposed work.

Vehicular Ad hoc Networks (VANETs) allow for vehicle-to-vehicle and vehicle-to-infrastructure communications using wireless local area network technologies. The distinctive features of their candidate applications (e.g., collision warning and local traffic information for drivers), resources (e.g., computational sources), and their ability to collect various data from their environment (e.g., vehicular traffic flow patterns) make VANETs a rich resource for information and resources. In this paper, we propose a new methodology to use VANETs to optimize signal control at traffic intersections as well as to create Vehicular Cloud (VC) computing environments. We theoretically analyze the traffic flow patterns in a given road intersection by using the diffusion approximation model. We calculate the probability of clearing the intersection and demonstrate the effect of the traffic patterns on the optimal choice of the traffic signal control parameters. Then, we employ our theoretical analysis to propose a potential solution to construct VANET-assisted VCs. Experimental results verify the correctness of our analysis.

For releasing the public concern on road safety, as an essential driving assistant technique for supporting autonomous deriving, considerable research efforts have been paid on developing practical traffic-related target/object detection methods. In recent years, by exploiting the powerful parallel processing capability of GPU and the feature extraction ability of deep convolutional neural network (CNN), the visible light image-based pedestrian detection method has gradually been considered as a potential solution. However, although it has been proven in the existing literature that CNN-based pedestrian detection methods can greatly improve the detection efficiency for lightly occluded pedestrians, the detection of low-visible pedestrians is still an open challenge. Accordingly, in this paper, we propose a novel collaborative pedestrian detection frame based on the Internet-of-Vehicles (IoV) to detect low-visible/hidden pedestrians or even hidden pedestrians. We further evaluate the proposed pedestrian detection framework relying on simulation experiments.

Reliable traffic flow prediction can greatly support the Intelligent Transportation System (ITS) to generate more effective traffic management decisions. Previous volume predictions mainly focused on the single road with simple flow patterns, such as suburban highways. However, with the development of the urban transportation system, the reliable flow information support becomes more significant for forming a solid ITS. Besides, travel delay is another widely neglected problem but can affect the prediction result significantly. Specifically, vehicles need some time to move from one place to another, and this time is called the travel delay. For further enhancing the prediction performance under the urban scenario, we propose a delay-based deep learning framework (MDGRU) to improve the accuracy of the short-term traffic flow prediction, in which travel delay is handled in the form of a weighted matrix enrolled into a multivariate input stacked Recurrent Neural Network (RNN). Multivariate input makes this approach has a stronger mining ability for spatial relationships capture, and the stacked structure leads to a more accurate pattern learning process. The results show that our approach is accurate and reliable.

Energy efficiency is a crucial issue for a practical wireless sensor network (WSN) due to the battery-powered sensors. For improving energy efficiency, many methods have been designed in WSNs thanks to the emerging techniques, i.e., data gathering and wireless charging. The data gathering algorithms are advantageous to decrease the energy consumption of nodes, and the wireless charging schemes can replenish energy to sensors for achieving the semi-permanent WSN. Though lots of approaches of each technique have been designed, the joint methods of both are still lacking. In this paper, a joint data gathering and wireless charging scheme is designed by adopting the mobile chargers (MCs) which can execute the energy charging and the data collection simultaneously. To decrease the data latency, an improved clustering algorithm is implemented first to construct the topology of the WSN. Then, a heuristic-based MC scheduling scheme is proposed for maximizing the charging utility while minimizing the energy consumption of MCs. Compared with the existing joint method, the proposed scheduling scheme achieves the outperformance on delay and charging utility.

Nowadays, wireless sensor networks (WSNs) are widely adopted by many civil/military applications. However, due to the limited capacity of the built-in battery, the lifetime of the sensor is limited, which in turn affects the working time of the whole system. Therefore, the limited energy supply is the most direct and critical constraint to maintain the long-term and efficient operation of the system. Accordingly, reducing energy consumption/improving energy efficiency is an essential prerequisite for designing a sustainable WSN. To address this problem, many approaches have been proposed. To help readers fully understand the techniques/methods in this area of research, we present a taxonomy of the existing energy-efficient strategies for achieving sustainable WSNs. We first introduce some basic concepts and assumptions commonly adopted in energy-efficient WSNs designs. Then, we discuss existing approaches designed for conventional WSNs (consisting of static nodes or nodes with limited mobility) from five aspects: clustering-based schemes, node deployment strategies, node scheduling algorithms, energy-efficient routing schemes, and energy-efficient joint designs. We compare these schemes and highlight their strengths and drawbacks. Additionally, we discuss state-of-the-art approaches relying on some emerging techniques, e.g., high-mobility data collectors, energy-harvesting techniques, etc. Finally, we conclude the paper and present some open challenges.

Recently, the Vehicular Network (VN) has received a lot of attention from researchers around the world. By allowing wireless communication, VNs enable information exchange among vehicles, which in turn has allowed drivers to become more aware of their surrounding road conditions. Accordingly, road safety is improved. However, due to the fast speed and frequent changes of direction of vehicles, the network topology of VNs is transient in nature. Hence, achieving efficient data dissemination/content delivery is a critical issue in the VNs-environment. In this article, we will introduce a novel passive roadside unit (RSU) detection-based proactive (PRDP) handover scenario. Consequently, the overhead of the handover process can be reduced, and the probability of successfully established connections can be improved. More precisely, by taking advantage of the Doppler effects of the received beacon signal, the passive RSU detection (PRD) scheme is derived by the maximum likelihood estimation function. Then, in combination with the extended Kalman filter (EKF), the PRDP handover protocol is designed to improve the energy efficiency of the handover procedure in the VNs-environment. We conduct intensive simulations to verify the proposed RSU detection scheme, and the experimental results further evaluate the performance of the proposed energy-efficient proactive handover protocol.

Nowadays, to deal with driving safety-related issues and improve travel comfort, the VehiculAr NETwork (VANET) has gained tremendous attention from researchers in both academia and industry around the world. By taking advantage of vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) communications, the VANET can significantly enhance road safety and travel comfort by improving drivers' awareness of their surrounding road environment and providing entertainment-related data service for passengers, respectively. However, due to the highly dynamic nature of the network topology in VANET, how to achieve reliable data transmission and content delivery is a critical task for implementing VANETs. Accordingly, in this article, we provide a novel data-highway-assisted content delivery protocol for addressing the content delivery problem in VANETs, in which, we explore the advantages of the predicted vehicular traffic volume driven by a newly designed fast traffic flow prediction scheme. We evaluate the performance of the proposed traffic flow prediction scheme by using three different data sets with different vehicles traffic flow patterns are chosen from the England Highways data set. Moreover, extensive simulations have been implemented to evaluate the proposed content delivery protocol.

Recently, to improve the efficiency and safety of the transportation system that is severely affected by the increasing traffic demand, the Internet-of-Vehicles (IoVs)/Vehicular Networks (VNets) have received more and more attention because it can effectively improve the ability of the participants in the transportation system to perceive the traffic environment around them through Vehicle-to-everything (V2X) technique. Moreover, V2X also makes it possible to share computing and storage power between vehicles, which further promotes the development of vehicular edge computing (VEC). However, due to the highly dynamic nature of the VNet's topology, based on the instant traffic flow condition, how to determine whether the vehicles on a given road can form a relatively stable cloudlet with certain computing or data storage capabilities to support certain VEC applications becomes a crucial task that needs to be solved. Therefore, in this paper, we proposed a Cloudlet-Dwell-time (CDT) estimation method to theoretically derive some essential parameters for implementing VEC applications, i.e., the vehicular cloudlet existence probability and its corresponding dwell-time. We further demonstrate the results of the proposed work based on traffic flow data chosen from the England Highways data set.

Due to the limited battery capacity, the lifetime and performance of the battery-powered WSNs are constrained. In order to prolong the lifetime, applying mobile data collectors to gather data in WSNs is a promising approach. In this paper, we design a two-phase data gathering strategy with the mobile data collector in the cluster-based WSN to improve energy efficiency and satisfy the delay constraints. More precisely, the sensors are divided into a set of clusters in the first phase, which ensures that the sensors can communicate with the mobile data collector within predetermined hops. We then develop the path for the mobile data collector using a genetic algorithm that is an applicable strategy for the optimization problem with respect to the shortest path finding in the large-scale WSNs. We evaluate the performance of the proposed path planning protocol by conducting intensive simulations. The simulation results indicate that the proposed scheme outperforms some state-of-the-art techniques on energy efficiency while enhancing the data update rate.

How to provide accurate and timely traffic flow information has become a hot topic in recent years since they can help schedule trips better and reduce traffic congestion. In previous studies, some machine learning (ML)-based models were proposed to predict the traffic volume at a single road segment/position, and these models performed not bad. However, when applied in a more complicated road network, they show low efficiency or need to pay higher computing costs. To solve this problem, an innovative ML-based model named selected stacked gated recurrent units model (SSGRU), is proposed for predicting the traffic flow through a sparse road network in this paper. There are mainly two parts in this model, one is used to do spatial pattern mining based on linear regression coefficients, and the other one includes a stacked gated recurrent unit (SGRU) which is essential for multi-road traffic flow prediction. A binarytree is adopted to approximate the sparse road network in the suburban area. To evaluate the proposed model, seven different traffic volume data sets recorded at 15-min interval are chosen from the England Highways data set to test our proposed work. The result shows that our model has greater adaptability and higher accuracy than others when applied to a multi-road input infrastructure.

In recent years, aiming to improve deriving safety and supporting autonomous driving, pedestrian detection has attracted considerable attention from both industry and academic. Moreover, by taking advantage of the powerful computational capacity of GPU and high-level feature learning ability of the deep convolutional neural network, tremendous image/video-based pedestrian detection methods have been proposed. However, most of the existing approaches are designed relying on the computer vision-based target detection techniques. Accordingly, the evaluation criteria they consider in the design are often from the computer vision research field. Therefore, these existing methods tend to focus on the improvement of accuracy and ignore some of the special requirements that need to be considered in the field of autonomous driving. In this paper, we will analyze and summarize the features of the state-of-the-art pedestrian detection methods in detail. Then, by considering the practical application scenarios of autonomous driving techniques, we further discuss the open challenges of designing a practical pedestrian detection method for supporting autonomous deriving.

In recent years, with the development of the Internet of Things (IoT) and the Vehicular Networks (VNets), a large number of computers and sensors equipped on different vehicles (e.g., onboard CPU, camera, GPS, etc.) can not only help the vehicle to collect its own surrounding environment information, but also share those information with other participants in the transportation system through Vehicle-to-everything (V2X) technique. This ability to share information further makes VNets a precious resource for information and resources, which can support the vehicular edge computing (VEC) environment. However, due to the high moving speed of vehicles and the relative motion between vehicles, the topology of vehicle networking is highly dynamic. How to estimate the number of vehicles and the time period that they can form a vehicular cloudlet in a road segment is a challenging task for enabling VEC. Hence, in the paper, we present a queueing model-assisted traffic density estimation scheme to derive and analyze some essential parameters for implementing VEC, i.e., the vehicular cloudlet existence probability and the corresponding lifetime. We further demonstrate the results derived by the proposed scheme.

Recently, the evolution of both wireless communication technologies and vehicular technology have greatly promoted the development of Vehicular Networks (VNs). The VN allows for vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) communications using wireless local area network technologies. The distinctive features of their candidate applications (e.g., collision warning and local traffic information for drivers), resources (e.g., computational sources), and their ability to collect various data from their environment (e.g., vehicular traffic flow patterns) make VNs a rich resource for information and resources. However, due to the transient nature of the network topology, data dissemination/content delivery is a challenging task in the VNs. Accordingly, in this article, we investigate the data dissemination/content delivery problem in VNs, and provide a novel traffic volume-aware data routing (TVDR) protocol for VNs. More precisely, by exploring the advantages of the various densities of coexistence vehicular traffic flows, the presented TVDR protocol can derive the data forwarding path with maximum link connection probability for the on-road vehicle. We evaluate the performance of the proposed TVDR protocol by conducting intensive simulations.

How to achieve a fast and safe data dissemination in the Vehicular ad-hoc network (VANET) is a hot research topic these days. However, the high mobility of the vehicles makes the topology of VANET unstable, and real-time road information is generally limited. Considering these shortcomings, it is helpful to use the accurate traffic prediction to assist the topology control in the VANET. For offering a better traffic flow prediction, this paper proposes an innovative hybrid prediction method, Delay-based Spatial-Temporal Autoregressive Moving Average model (DSTARMA) to enhance prediction effect. This model mainly focuses on dealing with the travel delay problem in short-term traffic flow prediction. In other words, vehicles always need some time to move from one place to another in a real traffic situation, and this period is called travel delay. In previous spatial-temporal models, no one takes this factor into account. In our model, the travel delay is handled in the form of spatial-temporal weighted matrices and treated as a key role. We evaluated our approach based on data in England highway traffic system. The result proves our approach is reliable and has the ability to offer more accurate road information in advance to support VANET.

Currently, traffic congestion has become a part of daily life of people in the cities around the world, and impacts people's lives adversely, e.g., the extra time spent on commuting, the extra exhaust emissions, etc. In order to reduce the effects of congestion on our lives, intensive research efforts have been proposed on this issue. Intelligent Transportation System (ITS) is one of potential solutions to enable various applications to improve road safety and travel comfort, and has gained a lot of attention from researchers around the world. In order to efficiently manage the transportation system and reduce traffic congestion, one of the paramount problems needed to be solved in ITS is the accurate traffic prediction. In this article, we firstly combine Fourier analysis with wavelet denoising technique to cope with the traffic flow forecasting problem. A two-layer fast Fourier transform (FFT)-based traffic prediction scenario is proposed, in which the discrete wavelet transform (DWT) with two different threshold values are adopted to decompose the high-frequent-noise and identify low-frequent traffic flow changing trend from the original data. Three different data sets with different traffic flow patterns are chosen from the England Highways data set to test our proposed work. Intensive simulations are implemented to verify the proposed work.

Wireless sensor networks (WSNs) have been widely adopted to fulfil the imperative requirement of real-time monitoring and/or long-term surveillance of the field-of-interest. However, due to the limited battery capacity, energy is the most critical constraint for improving the sustainability of a WSN. Hence, conserving energy and extending battery life are important in designing a sustainable WSN. Fortunately, the emerging energy harvest techniques provide us with a semi-permanent energy resource to power WSNs. In this article, we introduce a novel energy-aware hierarchical two-tier (HTT) energy harvesting-aided WSNs deployment scenario. More precisely, we consider two types of nodes in the system: one is the regular battery-powered sensor node (RSN), and the other is the energy harvesting-aided data relaying node (EHN). The objective is to use only RSNs to monitor FoI, while EHNs focus on collecting the sensed data from RSNs and forwarding the gathered data to the data sink. The minimum number of EHNs is deployed based on a newly designed probability density function to minimize the energy consumption of RSNs. This, in turn, extends the lifetime of the deployed WSN. The simulation results indicate that the proposed scheme outperforms some well-known techniques in the network lifetime, while enhancing the total throughput.

A wireless sensor network (WSN) is usually deployed in a field of interest (FoI) for detecting or monitoring some special events and then forwarding the aggregated data to the designated data center through sink nodes or gateways. Traditionally, the WSN requires the intensive deployment in which the extra sensor nodes are deployed to achieve the required coverage level. Fortunately, depending on the developments of the unmanned aerial vehicle (UAV) techniques, the UAV has been widely adopted in both military and civilian applications. Comparing with the traditional mobile sensor nodes, the UAV has much faster moving speed, longer deployment range and relatively longer serving time. Consequently, the UAV can be considered as a perfect carrier for the existing sensing equipment and used to form a UAV-based WSN (UWSN). In this paper, we theoretically analyse the coverage problem in the UWSN. Based on the integral geometry, we solve the aforementioned question. The experimental results further verifies our theoretical results.

The data dissemination and content delivery is a challenging research subject in the Internet of Things (IoT). Especially, in the Vehicular Networks environment, the network topology has the transient nature, due to the fast-moving velocity of the vehicles. One potential solution to this task is to improve the probability of the successful communication and content delivery. Hence, in this paper, we propose a novel and simple passive road side unit (RSU) localization scheme to estimate the location of the RSU, by which the vehicle can pre-determine the desired RSU to communicate based on its own position and routing information. By exploring the Doppler effects of the received signal, the RSU location estimator is derived by the maximum likelihood estimation method. Experimental results verify the correctness of the proposed estimation scheme.

A wireless sensor network (WSN) is usually deployed in a field of interesting (FoI) for detecting or monitoring some special events. Traditionally, the WSN requires intensive deployment in which the extra sensor nodes are deployed to achieve the required coverage level. While, depending on the developments of the unmanned aerial vehicle (UAV) techniques, the UAV has been widely adopted in both military and civilian applications. Comparing with the traditional mobile sensor nodes, the UAV has much faster moving speed, longer deployment range and relative long serving time. Consequently, the UAV can be considered as a perfect carrier for the existing sensing equipment and used to form a UAV-based WSN (UWSN). Naturally, in order to determine the efficiency of a UWSN, a question, “what is the probability of detecting a target in the Fol, if an UAV randomly scanned the FoI n times”, is raised. To solve this question, in this article, we theoretically analysed the target detection problem in the UWSN by considering static target and mobile target, respectively. The experimental results further verified our theoretical results.