loading page

GNN-based Deep Reinforcement Learning with Adversarial Training for Robust Optimization of Modern Tactical Communication Systems
  • Johannes Loevenich ,
  • Roberto Rigolin F. Lopes
Johannes Loevenich
Author Profile
Roberto Rigolin F. Lopes
Thales

Corresponding Author:[email protected]

Author Profile

Abstract

This paper investigates the feasibility of a Graph Neural Network (GNN)-based Deep Reinforcement Learning (DRL) for tackling complex optimization problems in modern communication systems deployed to tactical networks. Our methodology consists of three interacting agents: an environment builder agent responsible for generating complex network graph environments, a DRL agent situated within the control plane that possesses a global view of the current network state and makes decisions based on information gathered from various layers of the multi-layer tactical system, and an adversary designed to perturb the DRL agent, thereby evaluating its performance and robustness against data perturbations. Our numerical results indicate that enabling GNN in conjunction with adversarial training is crucial for the agent to learn the underlying network topology and parameters, ultimately enhancing the robustness of modern tactical communication systems operating in hostile environments.