Transport Layer Protocols Optimization for Satellite Networks(T4SAT)
bofreq-liu-transport-layer-protocols-optimization-for-satellite-networks-00

Document Type Declined BOF request
Title Transport Layer Protocols Optimization for Satellite Networks(T4SAT)
Last updated 2025-06-25
State Declined
Editor Yisong Liu
Responsible leadership
Send notices to (None)
bofreq-liu-transport-layer-protocols-optimization-for-satellite-networks-00

Name: Transport Layer Protocols Optimization for Satellite Networks (T4SAT)

Description

Existing transport protocols have performance challenges in LEO/MEO satellite networks. This is because LEO/MEO satellite networks differ significantly from terrestrial networks, and some of the design assumptions of TCP no longer hold in satellite environments. For example,
-High frequency handover: very often handover of communication connections between satellite to ground terminal and satellites when they are approaching the polar regions or Sun outage, etc.
-Variable data rates: satellite-to-ground link rate is highly impacted by atmospheric absorption, rain attenuation, scattering and diffraction, ionospheric scintillation and multipath effects.
-Variable link delay: the variation in RTT latency in satellite networks occurs at a higher frequency and with a larger magnitude, which presents significant challenges for TCP.

Measurements targeting Starlink indicate that there is significant room for performance improvement of transport layer protocols when operating in LEO satellite scenarios. Comparing the performance of UDP and TCP reveals that BBR only achieves about half of the maximum link capacity. To enhance the performance of transport protocols, the industry has undertaken a great deal of work, such as Multi-segment TCP, Multipath TCP, Transport Level Forward Error Correction (FEC), rate estimation, and Rate Pacing, etc.

Transport protocols need to achieve better throughput in LEO/MEO satellite scenarios, and must take into account factors such as high-frequency packet loss, highly variable bandwidth, and RTT (Round-Trip Time). Some possible enhancements are: efficient link bandwidth detection methods, enhanced congestion control that can distinguish between packet loss and congestion, and rapid packet loss detection and recovery methods. Meanwhile, the above enhancements need to ensure compatibility with existing transport layer protocols.

The BOF is supposed to identify and analyze current transport layer protocol challenges in LEO/MEO networks, while exploring potential work items and deliverables to address these technical issues and advance protocol optimization. Its primary objective is to assess whether sufficient interest exists within the IETF community to pursue collaborative work on the discussed topics.

Fill in the details below. Keep items in the order they appear here.

Required Details

Information for IAB/IESG

To allow evaluation of your proposal, please include the following items:
There was a side meeting on Transport Layer protocol optimization in satellite network, about 40 people attended that meeting and rough consensus about problems has been reached. The materials of this side meeting can be found in the link: https://github.com/Yisong2022/transport-layer-optimization-in-satellite-networks
This BoF will bring together interested parties, including those operators, researchers who have satellite network experiences, to discuss existing practices, possible solutions, what works, what does not work. Then identify potential avenues for IETF work to improve transport protocols performances on satellite network.

  • Any protocols or practices that already exist in this space: Transport layer protocols like TCP, QUIC
  • Which (if any) modifications to existing protocols or practices are required: To be identified with the ongoing discussions
  • Which (if any) entirely new protocols or practices are required: To be identified with the ongoing discussions
  • Open source projects (if any) implementing this work: None identified so far

Agenda

  • Administration: Chairs [5 minutes]
  • Use Cases [30 minutes]
  • 3GPP 5G NTN [15 minutes]
  • Video Services over Starlink [15 minutes]
  • Gap Analysis [15 minutes]
  • LEO Network Transport Layer Requirements [15 minutes]
  • Open Discussion [20 minutes]
  • Conclusions and Next Steps: Chairs and AD [5 minutes]