Computer Science Department, PhD Dissertation Defense Xiaokun Xu " Measuring and Improving Quality of Experience in Cloud Game Streaming"

Xiaokun Xu

PhD Candidate 

WPI – Computer Science Department 

Tuesday, December 16, 2025

Time: 10:00 a.m. – 11:00 a.m.

Zoom: https://wpi.zoom.us/j/3180706458

Committee members: 

Prof. Mark Claypool - Computer Science (Advisor)

Prof. Tian Guo - Computer Science

Prof. Lane Harrison - Computer Science

Prof. Carsten Griwodz, University of Oslo(External Member)

Abstract:

Cloud game streaming enables players to enjoy high-quality games without powerful local hardware by rendering content in the cloud and streaming it to lightweight clients. However, this approach is highly sensitive to network performance. Latency, jitter, and bandwidth variation can reduce responsiveness and visual smoothness, leading to degraded Quality of Experience (QoE).

This research systematically investigates how these factors influence cloud gaming QoE through controlled laboratory experiments and real-world in-home studies. By isolating the effects of delay, jitter, and visual degradation, the work identifies key predictors of player experience and develops mathematical models for accurately estimating QoE.

Beyond measurement and modeling, the dissertation also explores techniques for improving QoE across the system. At the network layer, it evaluates Active Queue Management (AQM) to stabilize delay and reduce jitter. At the client side, it analyzes playout buffer strategies to balance latency and smoothness. At the server side, it examines latency compensation techniques that help maintain fair and responsive gameplay when network delays cannot be avoided.

Together, these studies provide an integrated understanding of how QoE is shaped and offer practical guidance for improving cloud gaming performance across network, client, and server components.