-Reinforced Learning for NS3 in Cognitive Radio spectrum selection

-Spectrum Hole sensing/prediction is paramount importance to Secondary Users in a CR environment

-Problem of radio channel selection in a wireless multi-channel environment, e.g. 802.11 networks with external interference.

-RL optimisation algorithms can help predict the occurences of Spectrum White Space so that Channel can be effectively utilized

-The objective of the agent is to select for the next time slot a channel free of interference considering a basic example where the external interference follows a periodic pattern, i.e. time-slice sweeping over all available channels

WSL Ubuntu 18.04

NS3 3.17

Python3.6

GCC

TensorFlow 2.0

Keras

openpyxl

Copy folder under NS3-Gym / Scratch folder for execution

$ cd $NS3-GYM

$ ./waf --run "interference-pattern"

$ cd $NS3-GYM

$ cd scratch/interference-pattern

$ ./cognitive-agent-v3.py