NS3 Co-Simulation

[HaoRan Sun]

Hi everyone!

I have a idea about the NS-3 Co-Simulation with other environment(MATLAB or Python).

I want to make an simluation experimental scene, to simulating Multi-UAV multi-target tracking problem(Cooperative Multi-Robot Observation of Multiple Moving Targets).

Many Targets moving on the ground, many UAVs moving in the sky, each UAV has a certain target tracking  range and communication range, I want to design an UAVs moving algorithm that allows UAVs to cover as many ground targets as possible. 
In this scenario, the network communication between UAVs must be considered. Due to the problem of network transmission, the global map information obtained by a single UAV may be different from the actual situation. In this case, better moving decisions of UAV can be made.

My thinking is use NS3 to simulation network transmisson part, and use another to simulation the moving part, use TCP socket to connect.

I'm not sure if this option is possible, I really want to hear professional opinions and suggestion.

Thanks to discuss!

[Madan Kumar Pande]

Hi,

      This socket based connection has been done long ago for WiFi  module and AFIL, there was a paper published too.

      My two bits

With Regards,

madan

=================

--
Posting to this group should follow these guidelines https://www.nsnam.org/wiki/Ns-3-users-guidelines-for-posting
---
You received this message because you are subscribed to the Google Groups "ns-3-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ns-3-users+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/ns-3-users/13a305b0-4b3f-430f-9fb1-0b1d64eeb8d1n%40googlegroups.com.

[Théotime Balaguer]

Hi,

Co-simulation is possible with NS-3, however a quick review of the subject shows that this is technically more complexe than one might think. I also work with collaborative fleets of UAVs and I built a joint robotics and network co-simulator based on Gazebo (robotics) and NS-3 (network). My experience is that you should build your work upon on an existing co-simulation framework to overcome the basic co-simulation challenges that are synchronism and message exchange scheme.

I have not used it but the DEVS framework for co-simulation seems to be a good tool (take a look at this paper: 10.1177/0037549717749014).

On my side, I used the ROS-NetSim (10.1109/LRA.2021.3056347) framework, which seemed more straight forward and included the ROS framework.

Also, a more general advise is to keep in mind that the detail level of a simulation should always be chosen with regards to the research question you are asking. For instance, if you are only interested in the movements of the UAVs but are not interested in the control / robotics layers, maybe developing an advanced mobility model in NS-3 would be sufficient and way less painful than building a full co-simulation.

If you want more information, feel free to reach out to me. I have made a state-of-the-art review of co-simulation for multi-robot systems (sadly, it is written in French, I can translate it to English) and built my own.

Best regards,

Théotime