QoS-based multimetric routing protocols for vehicular networks
This master's thesis presents the design, implementation, and evaluation of a custom simulation based framework aimed at enhancing Quality of Service (QoS) in Vehicular Ad Hoc Networks (VANETs) through realistic mobility and communication models. VANETs are characterized by their high mobility...
| Author: | |
|---|---|
| Format: | master thesis |
| Publication Date: | 2025 |
| Country: | España |
| Institution: | Universitat Politècnica de Catalunya (UPC) |
| Repository: | UPCommons. Portal del coneixement obert de la UPC |
| Language: | English |
| OAI Identifier: | oai:upcommons.upc.edu:2117/450802 |
| Online Access: | https://hdl.handle.net/2117/450802 |
| Access Level: | Open access |
| Keyword: | Vehicular ad hoc networks (Computer networks)) VANETs Multimetric routing protocols Quality of Service OMNeT++ VEINS SUMO Weighted Round Robin Protocols d'encaminament multimètric Qualitat de Servei (QoS) WRR Xarxes vehiculars ad hoc (Xarxes d'ordinadors) Àrees temàtiques de la UPC::Enginyeria de la telecomunicació::Telemàtica i xarxes d'ordinadors |
| Summary: | This master's thesis presents the design, implementation, and evaluation of a custom simulation based framework aimed at enhancing Quality of Service (QoS) in Vehicular Ad Hoc Networks (VANETs) through realistic mobility and communication models. VANETs are characterized by their high mobility and dynamic topology, which pose significant challenges for traditional routing and communication protocols. These protocols often are not enough in ensuring timely and reliable data delivery under variable traffic conditions. To address these limitations, this work develops a custom simulation environment from scratch, integrating application level traffic differentiation and priority scheduling mechanisms designed for vehicular scenarios. The simulation framework is constructed using OMNeT++, SUMO, and VEINS, with all components configured for reproducibility and compatibility. A realistic urban mobility map and IEEE 802.11p standard based communication parameters were used to replicate real world conditions. The core contribution of this work is the implementation of a Constant Bit Rate (CBR) source application that generates four distinct traffic classes: beacons, emergency messages, traffic reports, and best-effort messages. These traffic types are scheduled using a custom Weighted Round Robin (WRR) mechanism, and mapped into IEEE 802.11p Access Categories (AC) allowing the network to prioritize critical data while preventing overload. To support intelligent packet forwarding, a neighbor table mechanism is introduced, enabling evaluation of nearby nodes based on distance and density. This multimetric approach provides a foundation for context aware routing decisions, improving the adaptability and responsiveness of the system. The simulation scenario uses random mobility traces within the urban map to test the scalability and effectiveness of the proposed design. Performance was assessed under two traffic density scenarios using key metrics such as end-to-end delay, packet delivery ratio (PDR), packet loss, and throughput. Results show that the proposed architecture performs in a stable manner and effectively manages traffic prioritization. However, limitations in route planning and mobility model realism impacted overall network efficiency. Longer and more realistic vehicle trips are identified as a necessary improvement for achieving better performance, particularly in terms of throughput and packet delivery. Despite this, the WRR scheduler demonstrated its ability to fairly distribute network resources among traffic classes. Additionally, the thesis includes a comprehensive appendix with all the installation and configuration steps of the environment. This guide covers Linux dependencies, OMNeT++, SUMO, and Veins versions, as well as deployment on a virtual machine hosted on a server. This documentation ensures that the framework can be reproduced and extended for future research. The effectiveness of the framework lies not only in its implementation but also in its modularity and adaptability. The combination of realistic traffic generation, priority based scheduling, and multimetric routing provides a robust platform for evaluating QoS strategies in VANETs. Even current performance metrics suggest the need for further optimizations, especially in forwarding techniques and vehicular mobility, the simulation environment remains functional and representative. In conclusion, this work offers a well structured and reproducible simulation framework that addresses key QoS challenges in vehicular networks. The proposed scheduling and routing strategies validate the potential of traffic aware and multimetric designs in improving network behavior. While there is room for improvement in terms of mobility models and dynamic routing, the current implementation provides a solid foundation for future works and real world applications. |
|---|