Simulation results demonstrate that the proposed CC-EVC technique improves the communication among vehicles by reducing channel congestion in a significant way compared to the Decentralized Congestion Control (DCC) technique.
Abstract
In Vehicular Ad-Hoc Networks (VANETs), each vehicle continuously broadcasts real-time information messages, such as speed, location, acceleration, and atmospheric conditions, to surrounding vehicles. However, in the current scenario, vehicle density increases continually on the road day by day, which is responsible for excessive amount of messages generation, leads to message congestion on channel and degrades the performance of VANETs. To mitigate this issue, proposed work investigates a novel technique named as Congestion Control with Enhancing Vehicle Connectivity (CC-EVC), empowered by machine learning (ML). The CC-EVC technique works with the principle of K-Mean algorithm for grouping the neighbouring vehicles to enhance connectivity. It restricts the communication of safety messages to non-emergency (outlier) vehicles to reduce the channel congestion and monitors the real-time channel load. The proposed technique also manages the channel congestion by grouping the neighbouring vehicles according to transmission range and using adaptive message transmission rate. The performance of the proposed CC-EVC technique is measured on SUMO tool by designing a dense vehicular network. The K-Mean technique is implemented using MATLAB, and performance parameters such as Packet Delivery Ratio (PDR), Normalized Routing Load (NRL), Throughput, and End-to-End (E2E) delay are evaluated using the NS2 simulator. Simulation results demonstrate that the proposed CC-EVC technique improves the communication among vehicles by reducing channel congestion in a significant way compared to the Decentralized Congestion Control (DCC) technique. Proposed CC-EVC technique achieves 89.7298% PDR, 111.73ms E2E delay, and 0.32 NRL with Ad-Hoc On-Demand Distance Vector (AODV) routing protocol and 82.447% PDR, 93.45ms E2E delay, and 0.67 NRL with Destination-Sequenced Distance Vector (DSDV) routing protocol at 500mWatt Transmission Power (TP) and 100bytes Packet Size (PS). Further, operational efficiency of the proposed work is compared with the existing AODV and DSDV protocols with varying TP and PS. It is demonstrated that the proposed technique results in enhanced throughput and PDR with AODV protocol whereas lowering the E2E delay and NRL with DSDV protocols.
Simulation results obtained demonstrate that Q-WeCBR outperforms CBR, DSDV, and GPSR in terms of packet delivery ratio and throughput, confirming the effectiveness of clustering combined with learning-based routing for dynamic vehicular networks.
Ahlam Boussadia· International journal of inf...· 0 citations
Multi-hop vehicular wireless networks are highly susceptible to channel congestion due to dynamic topologies, heterogeneous traffic densities, and contention-based medium access. This paper proposes a deployment-aware predictive congestion control framework for vehicular communications, where packet transmission decisions are guided by supervised machine learning models optimized under predictive and computational constraints. Its novelty lies in coupling packet-level congestion prediction with deployment-aware model selection, allowing transmission-control models to be selected according to predictive quality, measured native-inference latency, and serialized model footprint. A packet-level dataset was constructed from realistic vehicle trajectories generated with SUMO using road networks extracted from OpenStreetMap for three topologically diverse cities: Liverpool, Rio de Janeiro, and Houston. These trajectories support vehicular network evaluation under heterogeneous urban conditions, enabling the assessment of predictive transmission decisions across different road topologies. Several supervised learning models were considered, including linear classifiers, neural networks, and ensemble machine learning architectures such as Random Forest, Weighted Soft Voting, and Stacking. Hyperparameter tuning was formulated as a multi-objective optimization problem and solved using the NSGA-II evolutionary algorithm, jointly maximizing predictive performance while minimizing inference latency and serialized model size. Pareto-efficient models were exported to ONNX format and executed through a native inference pipeline to evaluate their suitability for computationally constrained vehicular communication environments. Results across heterogeneous urban scenarios show that the proposed framework improves vehicular communication performance in terms of packet delivery reliability and communication delay, while preserving competitive machine learning performance under deployment-oriented constraints.
Juan Pablo Astudillo León, Leticia Lemus Cárdenas, Luis J. de la Cruz Llopis et al.· IEEE Access· 0 citations
This article proposes a new approach to routing, termed RACER (Real-time Adaptive Congestion-aware Emergency Routing), which dynamically responds to changing traffic conditions without requiring additional traffic-signal-control infrastructure, relying instead on congestion information obtained through standard vehicle-to-infrastructure telemetry.
Harinath Ankarboina, Jasmini Kumari, Amit Kumar Singh et al.· IEEE Open Journal of the Com...· 0 citations
Due to increasing vehicle density, urbanization, and complex mobility patterns, road traffic injuries continue to pose a serious threat to public health and safety on a global scale. This ongoing crisis highlights the urgent need for intelligent, connected, and proactive vehicular systems that can prevent collisions, reduce injuries, and optimize traffic flow in real-time. The Internet of Vehicles (IoV) has become a key component of next-generation intelligent transportation, enabling seamless communication, data sharing, and collaborative decision-making among vehicles, roadside infrastructure, and cloud services. However, the effectiveness of current IoV communication frameworks in the real world is hampered by issues such as high latency, inefficient bandwidth utilization, limited scalability, and inadequate trust management. To address these challenges, this survey thoroughly examines 50 cutting-edge studies (2021–2025), including V2V (Vehicle-to-Vehicle), V2I (Vehicle-to-Infrastructure), and hybrid V2X (Vehicle-to-Everything) communication, edge–fog–cloud orchestration, 5G/6G integration, SDN (Software Defined Networking)/NFV (Network Function Virtualization) programmability, and security and trust-aware techniques. We provide a structured comparative analysis of communication types, enabling technologies, and limitations. Building on these insights, we propose an adaptive multi-tier IoV connectivity architecture that offers ultra-low latency, high scalability, and robust interoperability through distributed edge–cloud processing, AI-driven resource orchestration, adaptive blockchain-enabled security, and cross-technology communication control. Furthermore, we identify persistent research gaps and outline targeted future directions. The analysis suggests that AI-based optimization combined with hybrid and multi-tier designs has the potential to significantly improve network resilience, adaptability, and efficiency, offering a promising foundation for high-performance, secure, and reliable IoV systems.
Arbab Waheed Ahmad, M. Derawi, Raja Sana Gul· IEEE Open Journal of the Com...· 0 citations
Purpose: The study assessed the impact that machine learning algorithms, such as Gated Recurrent Unit and Naive Bayes, have on the performance of VANET.
Design/Methodology/Approach: Vehicular Ad hoc networks are formed by vehicles themselves, enabling communication between vehicles (V2V) and the roadside infrastructure (V2I), and allowing vehicles to accurately receive real-time information about the safety status of their surroundings and traffic flow. A dataset comprising network metrics, including timestamps for sent and received packets, average delay, and energy consumption, as well as performance metrics such as Packet Delivery Ratio, Throughput, and congestion state, was utilised to establish both input features and evaluation benchmarks in the study.
Research Limitation: Actual hardware implementation is difficult, which is a limitation
Findings: The baseline (GRU) was improved to (MGRU), i.e. modified GRU by +12.64%, and the baseline (NB) was improved to (MNB), i.e. modified Naïve Bayes by +18.33%. MGRU outperforms MNB, as MGU achieves high accuracy, high absolute throughput and packet delivery ratio, and low delay, reflecting better temporal modelling.
Practical Implication: The NS2 software was used to implement the VANET scenario.
Social Implication: This research will help in traffic jams and traffic congestion situations in vehicular communication.
Originality/Value: GRU was enhanced by modifying GRU and Naïve Bayes (NB) into Modified Naïve Bayes (MNB) to improve VANET.
M. Chawhan, N. Chandrashekhar, S. Gaigowal et al.· African Journal Of Applied R...· 0 citations