/HandoverOptimDRL

Framework for learning handover algorithms using deep reinforcement learning.

Primary LanguagePythonMIT LicenseMIT

HandoverOptimDRL

Framework for learning handover algorithms using deep reinforcement learning. This repo provides the codebase and results for [1].

Usage

  • The Vienna 5G Simulator routes can be found in the folder routes as .csv files.
  • The database used for the simulation is pre-stored in data_from_vienna. Use karlsruheScenario.m in the Vienna 5G Simulator to configure your scenario.
  • Take these files to start the Python script main.py. Please don't forget to create an output folder and adjust the file paths.

[1] Peter J. Gu, Johannes Voigt, and Peter M. Rost, "A Deep Reinforcement Learning-based Approach for Adaptive Handover Protocols in Mobile Networks"