FTRL-WRR: Learning-based two-path scheduler for LEO networks
| dc.contributor.author | Li, Daoping | |
| dc.contributor.supervisor | Pan, Jianping | |
| dc.date.accessioned | 2024-12-17T21:58:02Z | |
| dc.date.available | 2024-12-17T21:58:02Z | |
| dc.date.issued | 2024 | |
| dc.degree.department | Department of Computer Science | |
| dc.degree.level | Master of Science MSc | |
| dc.description.abstract | Multipath QUIC is inspired by the resource pooling principle, aiming to make a collection of resources behave as a single pool. However, current multipath schedulers tend to prioritize specific metrics like Round-Trip Time (RTT) or congestion window, often overlooking strategies that enhance overall resource usage and reduce flow completion time. This can lead to resource underutilization in high dynamic settings, such as those involving Low Earth Orbit (LEO) satellites. Addressing this challenge requires efficient traffic allocation to maximize bandwidth utilization. In this thesis, we verify that the relationship between traffic distribution and throughput in a two-path scenario resembles a quasi-concave function. Accordingly, we formulate the traffic allocation across two paths as a 1-dimensional optimization problem. To solve the two-path scheduling problem in dynamic environments, we introduce the FTRL-WRR algorithm. This approach integrates a Follow The Regularized Leader (FTRL) learner, ADWIN2 distribution change detector, and Weighted Round Robin (WRR) scheduler to enhance bandwidth utilization. We validate the effectiveness of the algorithm through extensive emulation and real-world testbed experiments, demonstrating consistent reduction in completion time across a range of scenarios. Additionally, we discuss the algorithm's limitations and suggest directions for future research. | |
| dc.description.scholarlevel | Graduate | |
| dc.identifier.uri | https://hdl.handle.net/1828/20854 | |
| dc.language | English | eng |
| dc.language.iso | en | |
| dc.rights | Available to the World Wide Web | |
| dc.subject | Two-path scheduling | |
| dc.subject | QUIC | |
| dc.subject | Starlink | |
| dc.subject | LEO satellite | |
| dc.subject | Bandit convex optimization | |
| dc.title | FTRL-WRR: Learning-based two-path scheduler for LEO networks | |
| dc.type | Thesis |