Abstract
The dynamic nature of traffic, high mobility and varying topology are major challenges to Urban Vehicular Ad Hoc Networks (VANETs), especially in densely populated urban areas. The methods of traditional routing and congestion control are not able to adjust and this leads to increased latency, decreased throughput and frequent breakdowns. In this paper, a new hybrid deep learning architecture that combines Graph Convolutional Networks (GCN) to extract spatial features and Long Short-Term Memory (LSTM) networks to model time is proposed with the aim of offering adaptive routing and other proactive congestion control capabilities in urban VANETs. The model applies the real time traffic information of sensors and vehicular communication systems in urban areas to forecast traffic congestion, model vehicle density and route optimization. In experimental results, the proposed GCN-LSTM model is significantly more efficient in routing and also in predicting results, compared to the baseline models. In particular, it recorded a Mean Absolute Error of 0.02, Root Mean Squared Error (RMSE) of 0.07, routing latency of 38.13 ms and Packet Delivery Ratio of 95, which was much better than GCN-only (MAE: 4.65, PDR: 90.3) and CNN-LSTM (MAE: 0.14, PDR: 88.1). The hybrid model also aids the real-time traffic processing through the edge-cloud cooperation, reducing the inference latency and scaling to city regions. Such methods as geospatial embedding and temporal batching enable the model to represent the intricate flow of traffic in high-resolution. The architecture is also modular and allows simple incorporation of both emergency vehicle prioritization and location-based rerouting and is scalable to operate on resource-constrained edge nodes, meaning it can be used in practice in intelligent city infrastructure. Altogether, the developed model offers a high-performing and scalable adaptive routing and congestion control solution to smart urban transportation systems.
Data availability
The dataset used in this study is the Urban Traffic Flow Dataset, publicly available on Kaggle [40]. It contains timestamped vehicle flow, speed, density, and congestion metrics collected from urban intersections using GPS and roadside sensors. The dataset supports high-resolution spatiotemporal analysis for traffic modelling and routing optimization.
References
Kanellopoulos, D., Sharma, V. K., Panagiotakopoulos, T. & Kameas, A. Networking architectures and protocols for Iot applications in smart cities: recent developments and perspectives. Electronics 12 (11), 2490. https://doi.org/10.3390/electronics12112490 (2023).
Medeiros, D. D. F., Souza, C. D., Carvalho, F. B. S. D. & Lopes, W. T. A. Energy-saving routing protocols for smart cities. Energies 15, 19, 7382. https://doi.org/10.3390/en15197382 (2022).
Silva, L. G. et al. Urban advanced mobility dependability: A model-based quantification on vehicular ad hoc networks with virtual machine migration. Sensors. 23 (23), (9485). https://doi.org/10.3390/s23239485 (2023).
Keeler, H., Błaszczyszyn, B. & Cali, E. Connectivity and interference in device-to-device networks in Poisson-Voronoi cities, arXiv: arXiv:2309.02137. https://doi.org/10.48550/arXiv.2309.02137 (2023).
Mazhar, T. et al. Quality of service (QoS) performance analysis in a traffic engineering model for next-generation wireless sensor networks. Symmetry 15 (2), 513. https://doi.org/10.3390/sym15020513 (2023).
Xu, W., Li, X., Hu, H., Wu & Li, J. Spatiotemporal analysis of urban road congestion during and post COVID-19 pandemic in Shanghai, China. Transp. Res. Interdisciplinary Perspect. 13, 100555. https://doi.org/10.1016/j.trip.2022.100555 (2022).
Ang, K. L. M., Seng, J. K., Ngharamike, E. & Ijemaru, G. K. Emerging technologies for smart cities’ transportation: geo-information, data analytics and machine learning approaches. ISPRS Int. J. Geo-Information. 11 (2). https://doi.org/10.3390/ijgi11020085 (2022).
Budda, R., Srinivasan, K., Chauhan, G. S., Jadon, R. & Gollapalli, V. S. T. Cloud-Integrated AI-Enhanced Software IoT Framework for Real-Time Water Quality Monitoring and E (Coli Prediction via Wireless Sensor Networks, 2024).
Prajapati, M., Upadhyay, A. K., Patil, D. H. & Dongradive, D. J. A review of deep reinforcement learning for traffic signal control. International J. Multidisciplinary Research.
Deshpande, N., Vaufreydaz, D. & Spalanzani, A. Behavioral decision-making for urban autonomous driving in the presence of pedestrians using Deep Recurrent Q-Network, in 16th International Conference on Control, Automation, Robotics and Vision (ICARCV), Shenzhen, China: IEEE, 428–433. https://doi.org/10.1109/ICARCV50220.2020.9305435 (2020).
Fuad, M. R. T. et al. Adaptive deep q-network algorithm with exponential reward mechanism for traffic control in urban intersection networks. Sustainability https://doi.org/10.3390/su142114590 (2022). ,14,21,14590.
Liu, X. Y., Zhu, M., Borst, S. & Walid, A. Deep reinforcement learning for traffic light control in intelligent transportation systems, arXiv: arXiv:2302.03669. (2023). https://doi.org/10.48550/arXiv.2302.03669
Paranjothi, A., Khan, M. S. & Zeadally, S. Survey on congestion detection and control in connected vehicles. Ad Hoc Netw. 108,102277 https://doi.org/10.1016/j.adhoc.2020.102277 (2020).
Giripunje, L. M., Vidyarthi, A. & Shandilya, S. K. Adaptive congestion prediction in vehicular ad-hoc networks (VANET) using Type-2 fuzzy model to Establish reliable routes, In Review. (2021). https://doi.org/10.21203/rs.3.rs-458059/v1
Gao, X., Haworth, J., Zhuang, D., Chen, H. & Jiang, X. Uncertainty quantification in the road-level traffic risk prediction by spatial-temporal zero-inflated negative binomial graph neural network(STZINB-GNN), arXiv: arXiv:2307.13816. (2023). https://doi.org/10.48550/arXiv.2307.13816
Farber, S. & Grandez, M. Transit accessibility, land development and socioeconomic priority: A typology of planned station catchment areas in the greater Toronto and Hamilton area. JTLU 10 (1). https://doi.org/10.5198/jtlu.2017.980 (2017).
Al, E. & Jyothi, E. V. N. A graph neural network-based traffic flow prediction system with enhanced accuracy and urban efficiency, jes, 19, 4, 336–349, (2024). https://doi.org/10.52783/jes.642
Mohammed, M. A. A comprehensive survey on congestion control techniques and the research challenges on VANET. Zanco J. Pure Appl. Sci. 34 (5), 50–78 (2022).
Kothai, G. et al. A new hybrid deep learning algorithm for prediction of wide traffic congestion in smart cities. Wirel. Commun. Mob. Comput. 1, 5583874. https://doi.org/10.1155/2021/5583874 (2021).
Mohanty, A. et al. Identification and evaluation of the effective criteria for detection of congestion in a smart City. IET Commun. 16, 5, 560–570. https://doi.org/10.1049/cmu2.12344 (2022).
Vallati, M. & Chrpa, L. Reducing traffic congestion in urban areas via real-time re-routing: a simulation study, in AI 2020: Advances in Artificial Intelligence,12576, M. Gallagher, N. Moustafa, and E. Lakshika, Eds., in Lecture Notes in Computer Science,12576. Cham: Springer International Publishing, 69–81. (2020). https://doi.org/10.1007/978-3-030-64984-5_6
Chin, Y., Bolong, N., Yang, S. S. & Teo, K. T. K. Q-Learning based traffic optimization in management of signal timing plan. Int. J. Simulation: Syst. Sci. Technol. https://doi.org/10.5013/IJSSST.a.12.03.05 (2020).
Anjaneyulu, M. & Kubendiran, M. Short-term traffic congestion prediction using hybrid deep learning technique. Sustainability 15 (1). https://doi.org/10.3390/su15010074 (2022).
Zhao, X., Hu, L., Wang, X. & Wu, J. Study on identification and prevention of traffic congestion zones considering Resilience-Vulnerability of urban transportation systems. Sustainability 14, 24, (16907). https://doi.org/10.3390/su142416907 (2022).
Zhong, W. & Du, L. Predicting traffic casualties using support vector machines with heuristic algorithms: a study based on collision data of urban roads. Sustainability 15 (4). https://doi.org/10.3390/su15042944 (2023).
Gao, F., Yang, B., Chen, C., Guan, X. & Zhang, Y. Distributed urban freeway traffic optimization considering congestion propagation, arXiv: arXiv:2106.06201. (2021). https://doi.org/10.48550/arXiv.2106.06201
Ghaffari, E., Rahmani, A. M., Saberikamarposhti, M. & Sahafi, A. An optimal path-finding algorithm in smart cities by considering traffic congestion and air pollution. IEEE Access. 10, 55126–55135. https://doi.org/10.1109/ACCESS.2022.3174598 (2022).
Zang,Jiao, J., Liu, S., Zhang, X., Song, G. & Yu, L. Identifying traffic congestion patterns of urban road network based on traffic performance index. Sustainability 15 (2). https://doi.org/10.3390/su15020948 (2023).
Urquiza Aguiar, L. F. & Aguilar Igartua, M. Contribution to the design of VANET routing protocols for realistic urban environments, Universitat Politècnica de Catalunya, (2016). https://doi.org/10.5821/dissertation-2117-98117
Dui, H., Zhang, Y., Zhang, S. & Zhang, Y. A. Recovery model and maintenance optimization for urban road networks with congestion, Mathematics, 11,9, (2023), (2004). https://doi.org/10.3390/math11092004
Xing, X., Li, X. & Zhai, Y. GRU-CNN neural network method for regional traffic congestion prediction serving traffic diversion demand. Wirel. Commun. Mob. Comput. 1–18. https://doi.org/10.1155/2022/8164105 (2022).
Zhai, Y., Wan, Y. & Wang, X. Optimization of traffic congestion management in smart cities under bidirectional long and short-term memory model. J. Adv. Transp. 1–8. https://doi.org/10.1155/2022/3305400 (2022).
Abdullah, S. M. et al. Optimizing traffic flow in smart cities: soft gru-based recurrent neural networks for enhanced congestion prediction using deep learning. Sustainability 15, 7, 5949. https://doi.org/10.3390/su15075949 (2023).
Tedjopurnomo, D. A., Choudhury, F. M. & Qin, A. K. TrafFormer: A transformer model for predicting long-term traffic, arXiv: arXiv:2302.12388. (2023). https://doi.org/10.48550/arXiv.2302.12388
Munuhwa, S. & Govere, E. Alleviating urban traffic congestion: case of Gaborone City. JESD https://doi.org/10.7176/JESD/11-8-06 (2020).
Rui, Y., Lu, W., Yi, Z., Wu, R. & Ran, B. A novel hybrid model for predicting traffic flow via improved ensemble learning combined with deep belief networks. Math. Probl. Eng. 1–16. https://doi.org/10.1155/2021/7328056 (2021).
Kempinska, K. & Murcio, R. Modelling urban networks using variational autoencoders. Appl. Netw. Sci. 4 (1), 114. https://doi.org/10.1007/s41109-019-0234-0 (2020).
Yan, L. & Wang, J. Deep reinforcement learning for ecological and distributed urban traffic signal control with multi-agent equilibrium decision making. Electronics https://doi.org/10.3390/electronics13101910 (2024). 13, 10, 1910.
Zhang, W. Countermeasures for urban traffic congestion in China from the perspective of system dynamics. Comput. Intell. Neurosci. 1–15. https://doi.org/10.1155/2022/3509902 (2022).
Mahmoud, A. & Mohammed, A. Leveraging hybrid deep learning models for enhanced multivariate time series forecasting. Neural Process. Lett. 56 (5), 223. https://doi.org/10.1007/s11063-024-11656-3 (2024).
Nazirova, S. et al. Secure data transmission in VANETs for philological field expeditions and mobile Language labs. JISIS 15, 2, 892–905. https://doi.org/10.58346/JISIS.2025.I2.059 (2025).
Wan, Y. Delay sensitive music transmission architecture for 5G VANET: integrated network slicing and predictive beamforming. Trans. Emerg. Telecommunications Technol. 36, 10, e70246. https://doi.org/10.1002/ett.70246 (2025).
Shi, X. Optimization of high-definition music streaming VANET broadcast protocol for QoS. Trans. Emerg. Telecommunications Technol. 36, 9e70248. https://doi.org/10.1002/ett.70248 (2025).
Ziya Urban Traffic Flow Dataset. Accessed: Aug. 13, 2025. [Online]. Available: https://www.kaggle.com/datasets/ziya07/urban-traffic-flow-dataset
Karim, A. A. & Nower, N. Long-term traffic prediction based on stacked GCN model. Knowl. Eng. Data Sci. 6 https://doi.org/10.17977/um018v6i12023p92-102 (2023).
Chu, K. F., Lam, A. Y. S., Tsoi, K. H., Huang, Z. & Loo, B. Y. Deep encoder cross network for estimated time of arrival. IEEE Access. 11, 76095–76107. https://doi.org/10.1109/ACCESS.2023.3294345 (2023).
Zhao, J. et al. Efficient and privacy-preserving federated learning against poisoning adversaries. IEEE Trans. Serv. Comput. 17, 5, 2320–2333. https://doi.org/10.1109/TSC.2024.3377931 (2024).
Paul, S. Investigating federated learning implementation challenges in 6G Network, in Fourth International Conference on Advances in Electrical, Computing, Communication and Sustainable Technologies (ICAECT), Bhilai, India: IEEE, 1–6, https://doi.org/10.1109/ICAECT60202.2024.10469650 (2024).
Pei, J., Liu, W., Li, J., Wang, L. & Liu, C. A review of federated learning methods in heterogeneous scenarios. IEEE Trans. Consumer Electron. 70, 3, 5983–5999. https://doi.org/10.1109/TCE.2024.3385440 (2024).
Acknowledgments
The Grantee is required to clearly acknowledge NCA and the provided grant in the “Funding Acknowledgment” section of any publication or patent resulting from the grant, using the following statement: “This research is supported by a grant (No. CRPG-25-3053) under the Cybersecurity Research and Innovation Pioneers Initiative, provided by the National Cybersecurity Authority (NCA) in the Kingdom of Saudi Arabia.”
Author information
Authors and Affiliations
Contributions
All contributions done by Dr. Maray.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
About this article
Cite this article
Maray, M. Hybrid deep learning techniques for adaptive routing and congestion control in urban VANET for wireless mobile networking. Sci Rep (2026). https://doi.org/10.1038/s41598-025-33193-2
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41598-025-33193-2
