Authors: Chang-Yong Lim, Jihong Park, Jinho Choi, Ju-Hyung Lee, Daesub Oh, Heewook Kim

Published on: February 04, 2024

Impact Score: 8.15

Arxiv code: Arxiv:2402.0235

Summary

• What is new: Introduction of the Ce2RACH protocol to mitigate inter-satellite interference in LEO satellite networks, achieving higher network throughput.
• Why this is important: Inter-satellite interference in downlink LEO satellite networks reduces the efficiency and throughput of these networks.
• What the research proposes: A multi-agent deep reinforcement learning framework that trains a more efficient multiple access protocol, Ce2RACH, which uses additional signaling messages for interference mitigation.
• Results: Ce2RACH demonstrated a 36.65% increase in network throughput compared to the existing eRACH protocol, though signaling message costs increased linearly with user numbers.

Technical Details

Technological frameworks used: Multi-agent deep reinforcement learning (MADRL)

Models used: Emergent Random Access Channel (eRACH), Centralized and Compressed Emergent Signaling for eRACH (Ce2RACH)

Data used: Simulations of downlink LEO satellite networks

Potential Impact

Satellite communication providers and manufacturers, telecommunication companies, and businesses reliant on satellite network infrastructure could significantly benefit, or need to adapt, with the adoption of Ce2RACH for improved network performance.

Want to implement this idea in a business?

We have generated a startup concept here: AstroNet Communications.