Public defence in Networking Technology, M.Sc. Xuebing Li

The title of the thesis: Enhancing Latency Reduction and Reliability for Internet Services with QUIC and WebRTC

Thesis defender: Xuebing Li
Opponent: Prof. Srinivasan Seshan, Carnegie Mellon University, US
Custos: Prof. Yu Xiao, Aalto University School of Electrical Engineering, Department of Information and Communications Engineering 

This dissertation presents innovative systems and algorithms designed to reduce latency and enhance the reliability of Internet services. To structure the research, the dissertation organizes the lifecycle of a service connection into three key stages: service discovery, delivery, and migration. 

A central focus of the study is the evaluation of QUIC and its potential to improve service discovery and migration. By integrating QUIC’s handshake process with traditional service discovery protocols, two systems are proposed to reduce connection setup latency and enhance the accuracy of end-user mapping, particularly by leveraging anycast and Domain Name System (DNS) technologies. Additionally, the solution extends QUIC’s handshake features, significantly improving the efficiency of service migration. 

In the realm of service delivery, the dissertation focuses on optimizing control mechanisms for real-time video streaming, using WebRTC as a testbed. It provides a detailed analysis of how control parameters impact streaming performance and application metrics, culminating in the development of an algorithm to optimize parameter settings during the slow-start phase of congestion control. Furthermore, a machine learning-based video streaming testbed is introduced, enabling AI models to jointly control multiple parameters, offering a more generalizable solution.

Key words: Service Discovery, Cross-layer Design, Real-time Video Streaming

Thesis available for public display 10 days prior to the defence at: https://aaltodoc.aalto.fi/doc_public/eonly/riiputus/

Contact:

Doctoral theses in the School of Electrical Engineering: https://aaltodoc.aalto.fi/handle/123456789/53