Introduction

This project is the code for our works:

Reinforcement Learning for Improved Random Access in Delay-Constrained Heterogeneous Industrial IoT Networks (under review)
Reinforcement Learning Random Access for Delay-Constrained Heterogeneous Wireless Networks: A Two-User Case (PDF)

Note that The code author of DLMA is YidingYu. We have made some adjustments for delay-constrained communication.

Run

If you want to run all algorithms in this project, you need to install some packets as follow:

The code of two-device case is in the folder "two_device".

You can run the algorithm by entering the corresponding folder and using the following command:

python main.py

And there is an example for how to calculate the transition probability matrix for "Upper Bound" in the folder "TransitionProbaility".

The code of two-device case is in the folder “multi_device”.

You can run the algorithm by entering the corresponding folder and using the following command:

python main.py

There are some arguments you can change for different settings.

For example:

python main.py -D 10 -n1 3 -n2 100

means that the deadline is 10, the number of uncontrollable devices is 3 and the number of TSRA is 100.

If you find this project useful, we would be grateful if you cite our paper.