Abstract
In this paper, we study the problem of optimizing the performance of vehicle-to-everything (V2X) using deep reinforcement learning techniques while sharing the spectrum between vehicle-to-infrastructure (V2I) links and vehicle-to-vulnerable road users (V2VRU) links in Cellular V2X (C-V2X). The objective is to protect VRU by improving the performance of V2VRU communications while maximizing the performance of V2I communications. Specifically, we formulate a spectrum sharing optimization problem with a two-objective function where the first objective is to improve the packet reception ratio (PRR) of VRU, whereas the second objective is to maximize the data rate of V2I communication links. To solve this challenging problem, we propose a deep reinforcement learning algorithm. A single agent controlling the vehicular network observes the environment and takes decisions accordingly by appropriately selecting the spectrum sub-bands and the transmission power levels. The simulation results show that the proposed scheme attains high performance compared to baseline solutions and solves the trade-off between maximizing the data rates of the vehicle users (V2I links) and improving the PRR of the V2VRU links.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
3rd Generation Partnership Project (3GPP): Technical Specification Group Radio Access Network; Study on enhancement of 3GPP Support for 5G V2X Services; (Release 15). Technical Report (TR), Version 15.1.0, (2017)
3rd Generation Partnership Project (3GPP): Technical Specification Group Radio Access Network; Study on LTE-based V2X Services; (Release 14). Technical Report (TR), Version 14.0.0, (2016)
Molina-Masegosa, R., Gozalvez, J.: LTE-V for Sidelink 5G V2X vehicular communications: a new 5G technology for short-range vehicle-to-everything communications. IEEE Veh. Technol. Mag. 12(4), 30–39 (2017). https://doi.org/10.1109/MVT.2017.2752798
Alalewi, A., Dayoub, I., Cherkaoui, S.: On 5G–V2X use cases and enabling technologies: a comprehensive survey. IEEE Access 9, 107710–107737 (2021). https://doi.org/10.1109/ACCESS.2021.3100472
Linget, T.: Vulnerable road user protection. 5GAA White Paper (2020)
Abid, M.A., Chakroun, O., Cherkaoui, S.: Pedestrian collision avoidance in vehicular networks. In: IEEE International Conference on Communications (ICC 2013), pp. 2928–2932. IEEE, Budapest (2013). https://doi.org/10.1109/ICC.2013.6654987
Rezgui, J., et al.: Deterministic access for DSRC/802.11p vehicular safety communication. In: 7th International Wireless Communications and Mobile Computing Conference, pp. 595–600. IEEE, Istanbul (2011). https://doi.org/10.1109/IWCMC.2011.5982600
Azizian, M., et al.: An optimized flow allocation in vehicular cloud. IEEE Access 4, 6766–6779 (2016). https://doi.org/10.1109/ACCESS.2016.2615323
Chakroun, O., et al.:Overhead-free congestion control and data dissemination for 802.11p VANETs. Veh. Commun. 1(3), 123–133 (2014). https://doi.org/10.1016/j.vehcom.2014.05.003
Rezgui, J., et al.: About deterministic and non-deterministic vehicular communications over DSRC/802.11p. Wireless Commun. Mobile Comput. 14(15), 1435–1449 (2014). https://doi.org/10.1002/wcm.2270
Azizian, M., et al.: Improved multi-channel operation for safety messages dissemination in vehicular networks. In: Proceedings of the fourth ACM International Symposium on Development and Analysis of Intelligent Vehicular Networks and Applications (DIVANet 2014), pp. 81–85. Association for Computing Machinery, New York (2014). https://doi.org/10.1145/2656346.2656410
Azizian, M., et al.: A distributed D-hop cluster formation for VANET. In: IEEE Wireless Communications and Networking Conference, pp. 1–6. IEEE, Doha (2016). https://doi.org/10.1109/WCNC.2016.7564925
Triwinarko, A., Cherkaoui, S., Dayoub, I.: Performance of radio access technologies for next generation V2VRU networks. In: IEEE International Conference on Communications (ICC 2022), pp. 1524–1529. IEEE, Seoul (2022). https://doi.org/10.1109/ICC45855.2022.9838580
Liang, L., Ye, H., Li, G.Y.: Spectrum sharing in vehicular networks based on multi-agent reinforcement learning. IEEE J. Sel. Areas Commun. 37(10), 2282–2292 (2019). https://doi.org/10.1109/JSAC.2019.2933962
Mlika, Z., Cherkaoui, S.: Network slicing for vehicular communications: a multi-agent deep reinforcement learning approach. Ann. Telecommun. 76(9), 665–683 (2021). https://doi.org/10.1109/JSAC.2019.2933962
Rashdan, I., de Ponte Muller, F., Wang, W., Schmidhammer, M., Sand, S.: Vehicle-to-Pedestrian channel characterization: wideband measurement campaign and first results. In: 12th European Conference on Antennas and Propagation (EuCAP 2018), 340 (5 pp.)-340 (5 pp.). Institution of Engineering and Technology (2018)
Mnih, V., et al.: Human-level control through deep reinforcement learning. Nature 518(7540), 529–533 (2015)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Triwinarko, A., Mlika, Z., Cherkaoui, S., Dayoub, I. (2023). Deep Reinforcement Learning to Improve Vehicle-to-Vulnerable Road User Communications in C-V2X. In: Sabir, E., Elbiaze, H., Falcone, F., Ajib, W., Sadik, M. (eds) Ubiquitous Networking. UNet 2022. Lecture Notes in Computer Science, vol 13853. Springer, Cham. https://doi.org/10.1007/978-3-031-29419-8_11
Download citation
DOI: https://doi.org/10.1007/978-3-031-29419-8_11
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-29418-1
Online ISBN: 978-3-031-29419-8
eBook Packages: Computer ScienceComputer Science (R0)