This project implements a vehicular communication system based on Fog Computing, developed and tested using the Eclipse MOSAIC 25.0 simulation framework. The system enables reliable, low-latency data exchange between vehicles (OBUs), Road-Side Units (RSUs), and a Fog node, with the ultimate goal of improving road safety.
For a better understanding of the project please read the article.
The aim of this project is to design and simulate a custom multihop forwarding protocol to ensure robust connectivity between vehicles and the Fog server. By leveraging ad hoc V2X communication and central Fog processing, the system is capable of:
- Creating a dynamic awareness map of neighboring vehicles,
- Optimizing forwarding paths based on proximity and hop count,
- Issuing real-time commands such as slow down, stop, or resume,
- Reacting to emergency vehicles in traffic scenarios.
- Eclipse MOSAIC 25.0 — Co-simulation framework for mobility and communication networks.
- SUMO (Simulation of Urban Mobility) — Used to simulate traffic behavior.
- Java — Main programming language used for OBUs, RSUs, and the Fog node logic.
- Ad hoc Networking (V2X) — Communication model for vehicles and RSUs.
.
├── article/ # Article diagrams and images
├── Network/ # Java source files for OBUs, RSUs, Fog node, and messages
├── scenarios/ # MOSAIC scenarios
├── compile.sh # Build script for the project
└── README.md
- Installed and configured Eclipse MOSAIC version 25.0
- Linux shell or WSL (for
compile.shscript)
-
Edit the build script
compile.sh: Set the following variables:BASE_DIR="/path/to/Network-Dir" TARGET_DIR="/path/to/eclipse-mosaic/scenario"
-
Compile the project:
./compile.sh
-
Run the scenario: From within the Eclipse MOSAIC root directory, execute:
./mosaic.sh -s vehApp_1
This will launch the simulation where regular vehicles yield to an emergency vehicle based on messages generated by the Fog.
- Dynamic neighbor map generation through Beacon and VehInfo messages
- Multihop routing based on proximity and hop optimization
- Emergency reaction through event-driven Fog messages
- Vehicle behavior adapted in real time in a SUMO traffic environment
The following improvements are planned:
- Integration of knowledge-based opportunistic forwarding algorithms.
- Evaluation and filtering of message freshness to prevent outdated data usage.
- Unified message handling via abstract classes (e.g.,
Messagesuperclass). - Statistical performance evaluation with metrics such as latency, throughput, and packet delivery using federation output.
Developed by:
- Augusto Campos (PG57510)
- Carlos Silva (PG57518)
Novos Paradigmas de Rede, Maio 2025