Comparative Analysis and Enhancement of Machine Learning Algorithms for Network Traffic Prediction in VANET

Advances in intelligent transportation systems: ML and DL techniques for traffic congestion forecasting

A key contribution of this study lies in its comparative synthesis of ML and DL models, revealing that hybrid and graph-based DL architectures consistently outperform traditional ML methods when handling large-scale, heterogeneous traffic datasets.

Leveraging an Optimized AI‐Driven Model for Estimating IoV Traffic Flow

Overall, the proposed Improved Dolphin Swarm‐optimized Dynamic Recurrent Neural Network shows promising potential for supporting intelligent traffic management and reducing traffic congestion; however, further validation using larger and more diverse datasets is required to confirm its generalizability and reliability.

MACHINE LEARNING ALGORITHMS IN NETWORK TRAFFIC CLASSIFICATION - A SURVEY

Effective NTC plays a vital role in enhancing bandwidth efficiency, ensuring network security, and maintaining Quality of Service (QoS) in todays communication systems. Conventional methods like port-based as well as payload-based classification are no more reliable because of the rise of encoded traffic and dynamic applications. As an alternative, Machine Learning (ML) approaches can automatically identify patterns in the statistical characteristics of network flows, offering more flexibility and accuracy. This paper presents a survey of Machine Learning that was used in NTC. The main focus on latest advances such as Protocol Detection, Traffic Behavior Analysis and Application Identification.Resent Trends, Evaluation Metrices, Benchmark Datasets and Machine Learning Algorithms are the core things that was discussed in this review. Along with Class Imbalance, Concept drift and encrypted traffic, these kinds of problem were also discussed in it. Beside these things, to understand the ML algorithms and make baseline benchmark as KDD dataset to represent comparative analysis.

Panoramic Review and Outlook of Traffic Flow Prediction Technology System

Deep learning-based multi-class misbehavior classification in fog-enabled vehicular ad hoc networks

Traffic collisions and congestion represent significant challenges within intelligent transportation systems (ITS). Consequently, a vehicular ad-hoc network (VANET) has been established. Numerous architectures have been incorporated into VANETs to manage the extensive data generated by vehicles. Collaboration with fog computing is vital, particularly for applications requiring real-time processing. In addition, there is a growing demand for advanced intrusion detection methods. These techniques are employed to identify the optimal response that the fog server should provide based on data received from a vehicle. As the network continues to grow, the amount of data needing analysis also increases. Thus, deep learning approaches become increasingly efficient. The requirement for feature selection is reduced when leveraging deep learning techniques. This work uses two deep learning-based misbehavior classification schemes for intrusion detection in VANETs: Long Short-Term Memory (LSTM) and Convolutional Neural Networks (CNN). The vehicular data that the Roadside Units (RSUs) obtain is initially sent to the fog server for preprocessing and classification. This study suggests four classification methods, which include 4 LSTM, 2 LSTM-2 CNN, 3 CNN-1 LSTM, and 2 CNN-1 LSTM within two main classifiers. The first classifier uses them to classify each type of vehicle, while the second uses them to classify generic behavior categories. The results of the first classifier were higher than those of the existing work. The precision ranges from 94% to 98%. The recall ranges from 92% to 97%. The F1 scores range from 94% to 97%. The second classifier also scores the best results compared to other works. The precision, recall, and F1-score range from 99% to 99.66%. The time and space complexity are calculated for each model in each classifier.

AI-Assisted Dynamic RSU Deployment Using Real-Time Traffic Density Estimation in VANETs

With the rapid advancement of vehicular communication technologies, maintaining reliable connectivity in Vehicular Ad Hoc Networks (VANETs) has become a critical challenge due to high mobility, dynamic topology, and uneven traffic distribution. Frequent disconnections in Vehicle-to-Vehicle (V2V) communication lead to increased latency and reduced network performance. To address these issues, this research proposes an AI-assisted dynamic Roadside Unit (RSU) deployment framework that leverages real-time traffic density estimation to optimize communication infrastructure. The proposed system utilizes deep learning-based vehicle detection models to analyze real-time traffic images and estimate vehicle density across different road segments. The extracted traffic information is further processed using machine learning techniques to predict communication demand and identify potential connectivity gaps. Based on these predictions, the system dynamically activates, deactivates, or repositions RSUs to ensure continuous network coverage and reduce dependency on unstable V2V links. The optimization model focuses on minimizing communication delay, enhancing packet delivery ratio, and improving overall network reliability through adaptive RSU placement. Additionally, a hybrid communication approach combining V2V and Vehicle-to-Infrastructure (V2I) is employed to overcome connectivity loss in sparse or highly dynamic traffic conditions. Simulation results demonstrate that the proposed AI-driven framework significantly improves network throughput, reduces communication latency, and ensures stable connectivity compared to traditional static RSU deployment strategies. The system effectively adapts to varying traffic patterns, making it suitable for next-generation intelligent transportation systems and smart city applications.