An open source Multi-Agent Deep Reinforcement Learning Routing Simulator for satellite networks

TL;DR

This paper presents an open source simulator for satellite network routing that compares traditional and advanced learning-based methods, demonstrating RL's advantages in reducing latency.

Contribution

It introduces a configurable, open source simulator supporting deep reinforcement learning for satellite network routing, with detailed visualization and analysis tools.

Findings

RL-based routing reduces end-to-end latency

Simulator supports multiple routing algorithms including Dijkstra and MA-DRL

Provides a flexible platform for satellite network research

Abstract

This paper introduces an open source simulator for packet routing in Low Earth Orbit Satellite Constellations (LSatCs) considering the dynamic system uncertainties. The simulator, implemented in Python, supports traditional Dijkstra's based routing as well as more advanced learning solutions, specifically Q-Routing and Multi-Agent Deep Reinforcement Learning (MA-DRL) from our previous work. It uses an event-based approach with the SimPy module to accurately simulate packet creation, routing and queuing, providing real-time tracking of queues and latency. The simulator is highly configurable, allowing adjustments in routing policies, traffic, ground and space layer topologies, communication parameters, and learning hyperparameters. Key features include the ability to visualize system motion and track packet paths. Results highlight significant improvements in end-to-end (E2E) latency using Reinforcement Learning (RL)-based routing policies compared to traditional methods. The source code, the documentation and a Jupyter notebook with post-processing results and analysis are available on GitHub.