CANE: A Cascade Control Approach for Network-Assisted Video QoE Management

Mehdi Hosseinzadeh, Karthick Shankar, Maria Apostolaki, Jay Ramachandran, Steven E. Adams, Vyas Sekar, Bruno Sinopoli

Research output: Contribution to journalArticlepeer-review

1 Scopus citations

Abstract

Prior efforts have shown that network-assisted schemes can improve the quality of experience (QoE) and QoE fairness when multiple video players compete for bandwidth. However, realizing network-assisted schemes in practice is challenging, as: 1) the network has limited visibility into the client players' internal state and actions; 2) players' actions may nullify or negate the network's actions; and 3) the players' objectives might be conflicting. To address these challenges, we formulate network-assisted QoE optimization through a cascade control abstraction. This informs the design of CAscade control-based NEtwork-assisted framework (CANE), a practical network-assisted QoE framework. CANE uses machine learning (ML) techniques to approximate each player's behavior as a black-box model and model predictive control (MPC) to achieve a near-optimal solution. We evaluate CANE through realistic simulations and show that CANE improves multiplayer QoE fairness by ∼ 50% compared with pure client-side adaptive bitrate (ABR) algorithms and by ∼ 20% compared with uniform traffic shaping.

Original languageEnglish (US)
Pages (from-to)2543-2554
Number of pages12
JournalIEEE Transactions on Control Systems Technology
Volume31
Issue number6
DOIs
StatePublished - Nov 1 2023
Externally publishedYes

All Science Journal Classification (ASJC) codes

  • Electrical and Electronic Engineering
  • Control and Systems Engineering

Keywords

  • Cascade control framework
  • fairness in quality of experience (QoE)
  • model predictive control (MPC)
  • multiplayer video streaming
  • network-assisted scheme
  • resource allocation

Fingerprint

Dive into the research topics of 'CANE: A Cascade Control Approach for Network-Assisted Video QoE Management'. Together they form a unique fingerprint.

Cite this