Awesome

MATLAB-NS3

Co-simulate MATLAB with NS-3 network simulator, combining the powers of MATLAB and NS-3. Describe the scenario in MATLAB, run simulation from MATLAB, capture the results and visualize them in MATLAB. Optionally use the MATLAB PHY and Channel models, instead of the statistical models of NS-3.

This co-simulation code currently:

• Supports modeling WLAN network and 802.11p based V2X scenarios.
  • See the documentation for the list of supported classes.
• Works with NS3.29 version (see the GitHub branch for NS3.26 version support)
• Works on Linux. Can get it working on Windows, if the NS3 library port is available to Windows.
• For high-fidelity WLAN PHY and Channel modeling, MATLAB WLAN Toolbox is recommended. Included example usage for 802.11a.
• For proper visualization of the included examples, display should be of Full-HD (1920x1080) resolution.

Example scenarios

Following example scenarios are included in this repository:

1. WLAN Infrastructure Example
2. Truck Platooning Example with V2X
3. Manhattan Grid Hazard Warning Example with V2X

See the included documentation in 'doc' folder, for more details.

Folders in the repository

• doc - Documentation about the co-simulation and included examples
• mlCode/mlWrappers - MATLAB wrappers over NS3
• mlCode/mlWiFiPhyChannel - Sample MALTAB code using MATLAB WLAN Toolbox Phy and Channel modeling
• native/mexBindings - Mex bindings over native C++ classes of NS3
• native/mlPhy - Additional C++ classes for NS3 to invoke MATLAB WLAN Toolbox Phy and Channel models
• scenarios - Contains example scenarios written in MATLAB that utilize the MATLAB-NS3 co-simulation
• videos - Recorded videos, expaining the demo scenarios

Setting up the co-simulation environment

Follow these steps to setup this co-simulation:

1. Download the source code (or download and extract the zip file). This gives out MATLAB-NS3 folder.

2. Download NS-3.29 from https://www.nsnam.org/release/ns-allinone-3.29.tar.bz2.

   • Untar and place NS-3 source folder 'ns-allinone-3.29' inside the MATLAB-NS3 folder
   • Execute ‘NS3-build.sh’ to build all the necessary NS3 libraries using the following command

   $sh NS3-build.sh

   Note: It is strongly recommended that you add LD_LIBRARY_PATH export to the shell startup script so that it is set automatically at startup.

3. Add path to 'LD_LIBRARY_PATH' in the startup script

   • If you use bash shell, add this line at the end of ~/.bashrc file:

   export LD_LIBRARY_PATH=<MATLAB-NS3 BASE FOLDER>/ns-allinone-3.29/ns-3.29/build/lib:$LD_LIBRARY_PATH

   • If you use csh / tcsh, add the following line at the end of ~/.cshrc file:

   setenv LD_LIBRARY_PATH<MATLAB-NS3 BASE FOLDER>/ns-allinone-3.29/ns-3.29/build/lib:$LD_LIBRARY_PATH

   Note: You have to launch new terminal (shell) or restart the PC to get the environment with updated LD_LIBRARY_PATH.

4. Launch MATLAB from the new terminal. Go to the folder MATLAB-NS3.

5. Run ‘buildMex.m’ to build Mex-binaries in the MATLAB Command Window.

   >>MATLAB-NS3/native/mexBindings/buildMex

Note: You may ignore warnings about GCC compiler version, thrown by MATLAB MEX. It works on Debian 8 (GCC version 4.9.x) and Debian 9 (GCC version 6.3.x). Should work with other GCC versions also.

Running the example scenarios

All examples are placed under ‘scenarios’ folder. Following steps are applicable to run any of the examples in this folder.

1. Go to any example folder, such as: MATLAB-NS3/scenarios/truckPlatooning/

2. Run ‘scenario.m’ script of any example. It might take around 2-3 minutes to complete. During this time the terminal will remain busy.

   Note: Running the simulation scenario will clear all the variables, close all the figures and unload all the functions in the current workspace.

   >>scenario.m

3. To visualize the results of the simulation, run the MATLAB script in the ‘visualizer’ folder of the same example.

   >>topologyVisualizer.m