QoS-oriented multipath protocol design in mobile networks
Date
2024
Authors
Yang, Wenjun
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Emerging applications demand stringent quality of services (QoS). Meanwhile, future networks are featured by ubiquitous mobility. How to meet users' QoS requirements in highly mobile environments remains an open issue, which motivates our research on QoS-oriented multipath transport layer protocol design in mobile networks.
First, multipath transfer is promising in tackling mobility issues for a seamless handoff. Scheduling packets across multiple paths, however, has the issue of out-of-order (OFO) arrival due to the heterogeneity of the paths. In this regard, we put forward a Mobility-Aware Multipath Scheduler (MAMS), ensuring that the reordering delay of each packet is minimized in various mobility scenarios and thus the QoS is significantly improved.
Enabling multipath transfer in the Integrated Terrestrial and LEO Satellite Network (ITSN) is promising. However, the existing multipath congestion control algorithms in ITSN suffer from bandwidth under-utilization or overshooting issues due to the high-speed network movement. Therefore, a novel Mobility-Aware COngestion control (MACO) algorithm is developed.
As applications are the driving force for protocol design, we investigate the performance of video streaming applications using multipath transfer. Assuming the QoS requirements of the application are known by the sender, we adopt a lightweight learning framework, a contextual multi-armed bandit (CMAB), to discover the underlying relationship between dynamic network states and QoS performance, which can intelligently select access networks and adapt FEC coding to trade off delay, reliability, and throughput.
Furthermore, 360-degree videos are not only bandwidth-intensive but also highly sensitive to delays. Ensuring both high video quality and smooth playback experience remains a critical issue. Therefore, we introduce a QoE-oriented Deadline-driven (RIDE) algorithm for multipath scheduling at the frame level. RIDE employs a dependency tree to understand deadlines for different types of frames and considers the negative impacts of Field of View (FoV) changes on scheduling decisions. Utilizing an actor-critic framework to train the neural network enables the scheduler agent to adapt to dynamic environments, including network and FoV dynamics.
Description
Keywords
QoS, mobility, multipath transfer