This repository aims to demonstrate the Constraint-Aware Deep Reinforcement Learning for vRAN Dynamic Placement implementation. In our tests, we used the Ubuntu Server 18.04 and Ubuntu 20.04, with IBM CPLEX solver, version 12.8, docplex library version 2.25.236, and Python version 3.8.10. All experiments were conducted on a computer server equipped with an Intel Core i7-10700F CPU @ 2.90GHz, 1 TB SSD, and 32 GB of memory.
In this paper, we introduce a Deep Reinforcement Learning (DRL) agent that is constraint-aware, ensuring the solutions' feasibility. We compare our DRL solution with existing optimization models and evaluate it under different scenarios, including the presence of Mobile Edge Computing (MEC) applications that compete for computing resources. Our contributions include a novel formulation, the implementation of a publicly available DRL agent, and insights into practical application scenarios for disaggregated vRAN optimization.
@inproceedings{almeida:24,
author = {Gabriel Almeida and Mohammad Abdel-Rahman and Kleber Cardoso},
title = { Constraint-Aware Deep Reinforcement Learning for vRAN Dynamic Placement},
booktitle = {Proceedings of the 42rd Brazilian Symposium on Computer Networks and Distributed Systems (SBRC 2024)},
location = {Niterói/RJ},
year = {2024},
keywords = {},
issn = {2177-9384},
pages = {337--350},
publisher = {SBC},
address = {Porto Alegre, RS, Brasil},
doi = {10.5753/sbrc.2024.1379},
url = {https://sol.sbc.org.br/index.php/sbrc/article/view/29803}
}