Traffic Data Research Articles

Wormhole attack detection and mitigation model for Internet of Things and WSN using machine learning.

The Internet of Things (IoT) is revolutionizing diverse sectors like business, healthcare, and the military, but its widespread adoption has also led to significant security challenges. IoT networks, in particular, face increasing vulnerabilities due to the rapid proliferation of connected devices within smart infrastructures. Wireless sensor networks (WSNs) comprise software, gateways, and small sensors that wirelessly transmit and receive data. WSNs consist of two types of nodes: generic nodes with sensing capabilities and gateway nodes that manage data routing. These sensor nodes operate under constraints of limited battery power, storage capacity, and processing capabilities, exposing them to various threats, including wormhole attacks. This study focuses on detecting wormhole attacks by analyzing the connectivity details of network nodes. Machine learning (ML) techniques are proposed as effective solutions to address these modern challenges in wormhole attack detection within sensor networks. The base station employs two ML models, a support vector machine (SVM) and a deep neural network (DNN), to classify traffic data and identify malicious nodes in the network. The effectiveness of these algorithms is validated using traffic generated by the NS3.37 simulator and tested against real-world scenarios. Evaluation metrics such as average recall, false positive rates, latency, end-to-end delay, response time, throughput, energy consumption, and CPU utilization are used to assess the performance of the proposed models. Results indicate that the proposed model outperforms existing methods in terms of efficacy and efficiency.

Predictive Modelling of Network Capacity Demands: A Machine Learning Approach for Global Enterprise Backbone Networks

In the dynamic landscape of global enterprise networks, accurate capacity forecasting is paramount for ensuring optimal resource allocation and preventing service disruptions. This paper presents a hybrid machine learning methodology that combines Autoregressive Integrated Moving Average (ARIMA) models with additional techniques to enhance the accuracy and reliability of network capacity forecasts. By leveraging historical traffic data and incorporating external factors, we develop a predictive model that outperforms traditional methods and adapts to the evolving demands of modern networks. The effectiveness of our approach is validated through rigorous testing against established benchmarks, demonstrating significant improvements in forecasting accuracy.

Dissipative Structure Properties of Traffic Flow in Expressway Weaving Areas

Expressway weaving areas meet dissipative structure characteristics. When traffic states reach a certain range, they exhibit self-organising criticality, and slight changes may trigger unpredictable congestion. This paper examines the correlation between the dissipative structure of the weaving area and key traffic parameters. The range of dissipative structure states in the weaving area is defined through the dissipative structure concept with three-phase traffic flow theory and real traffic data. Based on the fundamental diagram and measured traffic data, the weaving area dissipative structure model characterising the relationship between critical state changes in traffic volume is constructed and validated. Finally, the Cell Transmission Model simulation was used to examine the characteristic relationship between the weaving area dissipative structure state duration, the weaving area length and the weaving flow ratio. The results show that the length of the dissipative structure state is maintained when the traffic flow is self-organised into a free-flow or a congested state positively correlates with the length of the weaving area. Higher weaving flow ratios lead to shorter dissipative structure state durations during congestion formation, and the exact opposite during congestion evacuation. This paper is important for analysing the congestion mechanism and managing congestion.

Multi-Scale Spatio-Temporal Attention Networks for Network-Scale Traffic Learning and Forecasting.

Accurate and timely forecasting of traffic on local road networks is crucial for deploying effective dynamic traffic control, advanced route planning, and navigation services. This task is particularly challenging due to complex spatio-temporal dependencies arising from non-Euclidean spatial relations in road networks and non-linear temporal dynamics influenced by changing road conditions. This paper introduces the spatio-temporal network embedding (STNE) model, a novel deep learning framework tailored for learning and forecasting graph-structured traffic data over extended input sequences. Unlike traditional convolutional neural networks (CNNs), the model employs graph convolutional networks (GCNs) to capture the spatial characteristics of local road network topologies. Moreover, the segmentation of very long input traffic data into multiple sub-sequences, based on significant temporal properties such as closeness, periodicity, and trend, is performed. Multi-dimensional long short-term memory neural networks (MDLSTM) are utilized to flexibly access multi-dimensional context. Experimental results demonstrate that the STNE model surpasses state-of-the-art traffic forecasting benchmarks on two large-scale real-world traffic datasets.

A new spatiotemporal convolutional neural network model for short-term crash prediction

AbstractPredicting short-term traffic crashes is challenging due to an imbalanced data set characterized by excessive zeros in noncrash counts, random crash occurrences, spatiotemporal correlation in crash counts, and inherent heterogeneity. Existing models struggle to effectively address these distinct characteristics in crash data. This paper proposes a new joint model by combining the time-series generalized regression neural network (TGRNN) model and the binomially weighted convolutional neural network (BWCNN) model. The joint model aims to capture all these characteristics in short-term crash prediction. The model was trained and tested using real-world, highly disaggregated traffic data collected with inductive loop detectors on the M1 motorway in the UK in 2019, along with crash data extracted from the UK National Accident Database for the same year. The short-term is defined as a 30-min interval, providing sufficient time for a traffic control center to implement interventions and mitigate potential hazards. The year was segmented into 30-min intervals, resulting in a highly imbalanced data set with over 99.99% noncrash samples. The joint model was applied to predict the probability of a crash occurrence by updating both the crash and traffic data every 30 min. The findings revealed that 75.3% of crashes and 81.6% of noncrash events were correctly predicted in the southbound direction. In the northbound direction, 78.1% of crashes and 80.2% of noncrash events were accurately captured. Causal analysis and model-based interpretation were used to analyze the relative importance of explanatory variables regarding their contribution to crashes. The results reveal that speed variance and speed are the most influential factors contributing to crash occurrence.

Spatiotemporal Heterogeneity Analysis of Provincial Road Traffic Accidents and Its Influencing Factors in China

To objectively evaluate the road traffic safety levels across different provinces in China, this study investigated the spatiotemporal heterogeneity characteristics of macro factors influencing road traffic accidents. Panel data from 31 provinces in China from 2009 to 2021 were collected, and after data preprocessing, traffic accident data were selected as the dependent variables. Population size, economic level, motorization level, highway mileage, unemployment rate, and passenger volume were selected as explanatory variables. Based on the spatiotemporal non-stationarity testing of traffic accident data, three models, namely, ordinary least squares (OLS), geographically weighted regression (GWR), and geographically and temporally weighted regression (GTWR), were constructed for empirical research. The results showed that the spatiotemporal heterogeneity characterizing the macro factors of traffic accidents could not be ignored. In terms of impact effects, highway mileage, population size, motorization level and passenger volume had positive promoting effects on road traffic accidents, while economic level and unemployment rate mainly exhibited negative inhibitory effects. In terms of impact magnitude, highway mileage had the greatest impact on traffic accidents, followed by population size, motorization level, and passenger volume. Comparatively, the impact magnitude of economic level and unemployment rate was relatively small. The conclusions were aimed at contributing to the objective evaluation of road traffic safety levels in different provinces and providing a basis for the formulation of reasonable macro traffic safety planning and management decisions. The findings offer valuable insights that can be used to optimize regional traffic safety policies and strategies, thereby enhancing road safety.

Exploration of Traffic Accident-Based Pilot Zones for Autonomous Vehicle Safety Validation

Recently, the commercialization of autonomous vehicles has increased the importance of verifying vehicle safety through autonomous trials. Autonomous driving trials are conducted in limited areas within artificially constructed test roads and pilot districts and directly explore road sections and areas with similar environments to ensure the safety of AVs driving on real roads. Many previous studies have evaluated the complex response potential of AVs by deriving edge scenarios to ensure their safety. However, difficulties arise in exploring real roads with traffic accident factors and configurations similar to those in edge scenarios, making validation on real roads uncertain. This paper proposes a novel method for exploring pilot zones using traffic accident data to verify the safety of autonomous vehicles (AVs). The method employs a CNN + BiGRU model trained on DMV dataset data to classify traffic accidents as AV- or human-caused. The model’s classification accuracy was evaluated using recall, precision, F1 score, and accuracy, achieving 100.0%, 97.8%, 98.9, and 99.5%, respectively. The trained model was applied to the KNPA dataset, identifying 562 out of 798 cases as AV-like, indicating potential areas of high accident density due to AV operation. Outlier detection and DBSCAN clustering were utilized to identify compact pilot zones, effectively reducing the area size compared to raw data clusters. This approach significantly lowers the cost and time associated with selecting test roads and provides a viable alternative for countries lacking real AV accident data. The proposed method’s effectiveness in identifying pilot zones demonstrates its potential for advancing AV safety validation.

Fixed Automated stations location and UAVs routing problems in urban road Networks: A tailored Branch-&-price algorithm

Unmanned aerial vehicles (UAVs) serve as vital tools for collecting real-time traffic data, enabling rapid responses to events, and optimizing traffic management based on their cost-effectiveness, high flexibility, and wide coverage range advantages. The urban UAV cruising problem primarily focuses on selecting the locations of fixed automated station (FAS) and optimizing UAVs’ cruising routes. This paper introduces a strategy where UAVs can take off and land on different FASs to enhance urban UAVs’ cruising efficiency. Firstly, the problem is formulated as an integrated linear integer model (FU-ILP) which simultaneously determines the locations of FASs and the routes of UAVs. The objective is to minimize redundant cruising distance and cruising cycle time. The FU-ILP model takes into account UAV‘s flight capabilities, FAS’s signal coverage range, power requirements, and battery charging. Moreover, we design a tailored branch-and-price (TB&P) algorithm to decompose the FU-ILP model, which can reduce the variable scale for solving larger-scale cases. Finally, we test the model and the algorithm on the Sioux Falls Test Network. The TB&P algorithm improves computational efficiency by nearly 80%. Moreover, our strategy yields superior results compared to the strategy where UAVs must return to their take-off FASs for recharging. We also demonstrate that increasing the number of UAVs does not reduce redundant cruising distance in the whole network, but it can shorten the cruising cycle. These policy conclusions also apply to the case of Xiongan. This provides a basis for managers to calculate the minimum UAV’s configuration based on the cities’ actual cruising frequency requirements.

Enhancing Traffic Control with AI Blockchain and Dynamic Computation Techniques

The rapid urbanization and increasing vehicular density in modern cities have led to significant challenges in traffic management and control. As urban areas continue to expand, the demand for more efficient and intelligent traffic control systems has become increasingly critical. This paper presents a novel approach to enhancing traffic management by integrating Artificial Intelligence (AI), Blockchain technology, and Dynamic Computation Techniques. AI is utilized to analyze and predict traffic patterns, enabling real-time adjustments to traffic signals and flow management. Blockchain provides a secure, transparent, and decentralized platform for data sharing and coordination among various stakeholders, ensuring data integrity and trust. The incorporation of Dynamic Computation Techniques allows for flexible and scalable processing of complex traffic data, facilitating rapid decision-making and adaptation to changing conditions. This multidisciplinary approach not only improves traffic efficiency and reduces congestion but also paves the way for more resilient and sustainable urban transportation systems. The findings highlight the transformative potential of combining AI, Blockchain, and advanced computation methods in the field of traffic control.

Spatial-Temporal Transformer Networks for Traffic Flow Forecasting Using a Pre-Trained Language Model.

Most current methods use spatial-temporal graph neural networks (STGNNs) to analyze complex spatial-temporal information from traffic data collected from hundreds of sensors. STGNNs combine graph neural networks (GNNs) and sequence models to create hybrid structures that allow for the two networks to collaborate. However, this collaboration has made the model increasingly complex. This study proposes a framework that relies solely on original Transformer architecture and carefully designs embeddings to efficiently extract spatial-temporal dependencies in traffic flow. Additionally, we used pre-trained language models to enhance forecasting performance. We compared our new framework with current state-of-the-art STGNNs and Transformer-based models using four real-world traffic datasets: PEMS04, PEMS08, METR-LA, and PEMS-BAY. The experimental results demonstrate that our framework outperforms the other models in most metrics.

A Bayesian tensor ring decomposition model with automatic rank determination for spatiotemporal traffic data imputation

Recently, tensor factorization models have shown superiority in solving traffic data imputation problem. However, these approaches have a limited ability to learn traffic data correlations and are easy to overfit when the pre-defined rank is large and the available data is limited. In this paper, we propose a Bayesian tensor ring decomposition model, utilizing Variational Bayesian Inference to solve the model. Firstly, tensor ring decomposition with an enhanced representational capability is used to decompose partially observed data into factor tensors to capture the correlation in traffic data. Secondly, to address the issue of selecting large pre-defined rank when data availability is limited, an automatic determination mechanism of tensor ring ranks is proposed. This mechanism can be implemented by pruning the zero-component horizontal and frontal slices of the core factors in each iteration, reducing the dimensions of the core factors and consequently lowering the tensor ring ranks. Finally, extensive experiments on synthetic data and four diverse types of real-world traffic datasets demonstrate the superiority of the proposed model. In the Guangzhou dataset, the maximum improvement in Mean Absolute Percentage Error can reach 15 % compared to the most competitive baseline model.

Real-time traffic condition uncertainty quantification using adaptive grey prediction interval model

Uncertainty quantification is important for making reliable transportation decisions. For grey-based uncertainty quantification approaches, the data classification methods for most models cannot yield real-time upper and lower limited data sequence, limiting their application in dynamic transportation systems. Therefore, this paper proposes an adaptive grey prediction interval model to quantify real-time traffic condition uncertainty. To this end, polynomial regression is first used to fit the traffic flow trend function in real time, generating dynamically the upper and lower limited data sequence. Then, upper grey (UG) and lower grey (LG) models are built with parameters optimised using particle swarm optimisation. Finally, based on the real-time upper and lower bounds generated by the UG and LG models, prediction intervals are constructed. Using real-world traffic flow data, the proposed model was shown to be able to generate workable prediction intervals in real time. Future studies are recommended to advance traffic condition uncertainty quantification.

Multi-view multi-label network traffic classification based on MLP-Mixer neural network

Network traffic classification is the basis of many network security applications and has received significant attention in the field of cyberspace security. Existing research on deep traffic analysis typically involves converting traffic data into images to extract spatial traffic features using Convolutional Neural Networks (CNNs). However, this approach ignores the semantic differences and details in the various packet structures. In this paper, we propose an MLP-Mixer based multi-view multi-label neural network for network traffic classification. Compared with the existing CNN-based methods, our method adopts the MLP-Mixer structure, which is more in line with the structure of the packet than the conventional convolution operation. In our method, one packet is divided into the packet header and the packet payload, together with the flow statistics of the packet as input from different views. We utilize a multi-label setting to learn different scenarios simultaneously to improve the classification performance by exploiting the correlations between different scenarios. We conduct experiments on three public datasets, and the experimental results show that our method can achieve superior performance. Code is available at https://github.com/ZxuanDang/MV-ML-traffic-classification.

Impact of flexible working on traffic congestion in extreme weather conditions: Empirical evidence from a natural experiment

IntroductionClimate change increases the frequency and intensity of extreme weather events, which can lead to increased congestion on roadways as drivers slow down and navigate hazardous conditions. Flexible working can be a potential solution because it reduces the need for employees to commute to work. MethodsThis study investigates the traffic congestion spatiotemporal patterns of flexible working in extreme weather events (i.e., heavy snow) by using the public health emergency as a natural experiment. We collect real-time traffic data from Harbin, China, and provide a framework to quantify the reduction of traffic congestion under extreme weather conditions. During the epidemic lockdown period, only crucial workers were allowed to go to work in the study area; everybody else was working from home. This is the maximum level of flexible working that a system can allow. Hence, our findings provide an upper limit for traffic congestion reduction in extreme weather events. We constructed three scenarios, i.e., baseline, snow, and snow with work-from-home (WFH). We use time series analysis and Kaplan-Meier survival analysis methods to study the spatiotemporal patterns of traffic congestion during morning peak hours (6:30 a.m.–9:30 a.m.) at 10-min intervals. ResultsThe data analysis identified that significant traffic congestion reduction due to the WFH arrangement. For example, the longest travel duration is reduced from 120 min in the snow scenario to 50 min in the snow with WFH scenario. ConclusionsThis study reveals the geographical patterns of urban traffic congestion, providing support for guiding residents to optimize snowy travel methods in future interventions, policy changes, and research.

Internet of things and optimized knn based intelligent transportation system for traffic flow prediction in smart cities

The rapid expansion of urban areas and the increasing number of vehicles on the road have resulted in accidents, traffic congestion, economic repercussions, environmental deterioration, and excessive fuel consumption. A dependable traffic management system is necessary to anticipate and regulate urban traffic patterns. Traffic forecast aids in the prevention of traffic issues. Urban traffic predictions often utilise historical and current traffic flow data to forecast road conditions. This article presents a traffic flow prediction system that utilises the Internet of Things (IoT), machine learning, and feature selection. Internet of Things (IoT) devices located on highways or within cars gather sensor data in real-time. The input data set comprises both real-time Internet of Things (IoT) data and historical traffic statistics. The input data is stored in a centralized cloud. The data is subjected to preprocessing in order to eliminate any unwanted interference and identify any exceptional values. The accuracy and root mean square error are contingent upon the process of feature selection. Particle swarm optimization identifies and extracts crucial features from input data. The classification model is constructed using K Nearest Neighbor, Multi layer Perceptron, and Bayes network approaches. The UCI traffic data is used for conducting experiments. The dataset has 47 attributes and 2102 occurrences. The accuracy of traffic flow prediction using PSO KNN is 96 %. The PSO KNN algorithm achieved a Mean Square Error (MSE) of 0.00289 and a Root Mean Square Error (RMSE) of 0.0595.

On the utility of probabilistic closed-form proxy models for describing supercomputer network traffic data

On the utility of probabilistic closed-form proxy models for describing supercomputer network traffic data

Optimal Feature Picking for Intrusion Detection on The Basis of Explainable Artificial Intelligence

Traditional security mechanisms for intrusion detection are strong, and the use of machine learning algorithms to detect intrusion into IoT networks has increased. As a result, the efficiency of the IoT network has increased and an effective model can be developed to detect attacks on the network. UNSW-NB15 is an IoT-based publicly available dataset for traffic data that contains general activity and malicious activity. Using this dataset, the key features were selected and these features fed different classifiers for training and classifying the attack in the network. Ann model performance levels of accuracy, the ML classifier, and accompanying algorithms for constructing a digital security system based on packet flow in-network, and features that really can monitor anomalous botnet behaviors proven to be highly efficient. The decision-making model for the classification was visualized using proven Explainable AI (XAI) approaches utilizing Scikit-Learn, LIME, ELI5, and SHAP libraries to boost explainability in classification predictions. The findings showed that XAI is realistic and profitable since security expertise and specialists stand to benefit greatly from combining conventional machine learning methods with Explainable AI (XAI) methodologies.

Dynamic attention aggregated missing spatial–temporal data imputation for traffic speed prediction

Traffic speed forecasting is indispensable in Intelligent Traffic Systems (ITS). The missing traffic speed due to sensors’ failure may hamper the accurate prediction of traffic speed. Different kinds of imputation techniques were employed to replace the missing values considering the spatial and temporal information. However, previous techniques did not consider the dynamic contribution of the neighboring nodes and failed to capture the complex properties of traffic speed data precisely. The architecture of the previous imputation models was infeasible in real-time applications in terms of scalability. Therefore, a novel machine learning-based imputation technique is proposed to generate the missing traffic speed using the spatial–temporal information and dynamic contribution of neighboring nodes. A new dataset considering the characteristics of the traffic speed data is generated from the existing dataset by exploiting the non-missing traffic speed of neighboring nodes. The newly formed dataset helps in the estimation of the dynamic contribution of the neighboring nodes using the linear regression algorithm. The non-missing speeds and estimated dynamic contribution of neighboring nodes assist in calculating the missing speeds of a node. Furthermore, a novel deep learning-based prediction model is introduced to forecast traffic speed accurately. The prediction model contains graph structure learning, a deep graph neural network with skip connections, GRU, multi-head soft attention, and a fully connected neural network. Graph structure learning is proposed based on the distribution of the congestion to represent the graph in an efficient way. The deep graph neural network with skip connections and GRU help in capturing topological and temporal information, respectively. The multi-head soft attention is designed to focus on every relevant temporal information at each time step for capturing global traffic information. Finally, the extracted spatial–temporal features are fed into the fully connected neural network to predict the traffic speed. The proposed imputation and prediction models are evaluated on two real-time datasets, METR-LA and PeMSD7 datasets, under different missing rates of different missing patterns. The proposed framework generates possible spatial–temporal information for efficient traffic congestion management and evicts erroneous issues in real-time.

Two-step dynamic obstacle avoidance

Dynamic obstacle avoidance (DOA) is a fundamental challenge for any autonomous vehicle, independent of whether it operates in sea, air, or land. This paper proposes a two-step architecture for handling DOA tasks by combining supervised and reinforcement learning (RL). In the first step, we introduce a data-driven approach to estimate the collision risk (CR) of an obstacle using a recurrent neural network, which is trained in a supervised fashion and offers robustness to non-linear obstacle movements. In the second step, we include these CR estimates into the observation space of an RL agent to increase its situational awareness. We illustrate the power of our two-step approach by training different RL agents in a challenging environment that requires to navigate amid multiple obstacles. The non-linear movements of obstacles are exemplarily modeled based on stochastic processes and periodic patterns, although our architecture is suitable for any obstacle dynamics. The experiments reveal that integrating our CR metrics into the observation space doubles the performance in terms of reward, which is equivalent to halving the number of collisions in the considered environment. We also perform a generalization experiment to validate the proposal in an RL environment based on maritime traffic and real-world vessel trajectory data. Furthermore, we show that the architecture’s performance improvement is independent of the applied RL algorithm.

Transfer learning-based nonstationary traffic flow prediction using AdaRNN and DCORAL

Traffic flow prediction is an integral part of an intelligent transportation system (ITS) for proactive transportation planning and management in public transit network systems. However, due to the influence of passengers’ travel modes and the difficulty of data collection, some traffic flow datasets are not only nonstationary in temporal distribution but also insufficient in data size, leading to inaccurate prediction results. Therefore, a transfer learning-based method is proposed to predict nonstationary traffic flow with inadequate data in the public transit system. In the proposed model, a novel approach named AdaRNN-DCORAL is presented to achieve accurate traffic flow prediction in the temporal dimension. Specifically, AdaRNN is an adaptive prediction network for nonstationary traffic flow data, while DCORAL performs domain adaptation between the source and target domain in the public transit system. First, two domains of traffic flow datasets in two public transit modes are established, which are the target domain with insufficient data and the source domain with sufficient data. Subsequently, the traffic flow datasets are divided into weekday and holiday time series based on their distribution characteristics. On this basis, AdaRNN-DCORAL is trained on the constructed datasets, thus achieving the precise nonstationary prediction of data scarcity in the target domain. Finally, the experiments demonstrate the improved performance of the proposed method over that of state-of-the-art models in terms of prediction accuracy and computation time.