Game of streaming players: Is consensus viable or an illusion?

Name: Game of streaming players: Is consensus viable or an illusion?
Author: Bentaleb, A., Beğen, Ali Cengiz, Harous, S., Zimmermann, R.

İsim	Game of streaming players: Is consensus viable or an illusion?
Yazar	Bentaleb, A., Beğen, Ali Cengiz, Harous, S., Zimmermann, R.
Basım Tarihi:	2019-08
Basım Yeri	- Association for Computing Machinery, Inc
Konu	HAS, QoE, ABR, Scalability, Instability, Unfairness, Underutilization, Game theory, Consensus theory
Tür	Süreli Yayın
Dil	İngilizce
Dijital	Evet
Yazma	Hayır
Kütüphane:	Özyeğin Üniversitesi
Demirbaş Numarası	1551-6857
Kayıt Numarası	e57cdc13-cefa-4a6f-9e36-daf3ab1fc760
Lokasyon	Computer Science
Tarih	2019-08
Notlar	Singapore Ministry of Education Academic Research Fund Tier 1 under MOE's grant ; UAE University
Örnek Metin	The dramatic growth of HTTP adaptive streaming (HAS) traffic represents a practical challenge for service providers in satisfying the demand from their customers. Achieving this in a network where multiple players share the network capacity has so far proved hard because of the bandwidth competition among the HAS players. This competition is exacerbated by the bandwidth overestimation that is introduced due to the isolated and selfish behavior of the HAS players. Each player strives individually to select the maximum bitrate without considering the co-existing players or network resource dynamics. As a result, the HAS players suffer from video quality instability, quality unfairness, and network underutilization or oversubscription, and the players observe a poor quality of experience (QoE). To address this issue, we propose a fully distributed game theory and consensus-based collaborative adaptive bitrate solution for shared network environments, termed Game Theory and consensus-based Approach for Cooperative HAS delivery systems (GTAC). Our solution consists of two-stage games that run in parallel during a streaming session. We extensively evaluate GTAC on a broad set of trace-driven and real-world experiments. Results show that GTAC enhances the viewer QoE by up to 22%, presentation quality stability by up to 24%, fairness by at least 31%, and network utilization by 28% compared to the well-known schemes.
DOI	10.1145/3336496
Cilt	15

Kaynağa git Özyeğin Üniversitesi

Aramaya Dön

Özyeğin Üniversitesi

Kaynağa git

Game of streaming players: Is consensus viable or an illusion?

Yazar Bentaleb, A., Beğen, Ali Cengiz, Harous, S., Zimmermann, R.

Basım Tarihi 2019-08

Basım Yeri - Association for Computing Machinery, Inc

Konu HAS, QoE, ABR, Scalability, Instability, Unfairness, Underutilization, Game theory, Consensus theory

Tür Süreli Yayın

Dil İngilizce

Dijital Evet

Yazma Hayır

Kütüphane Özyeğin Üniversitesi

Demirbaş Numarası 1551-6857

Kayıt Numarası e57cdc13-cefa-4a6f-9e36-daf3ab1fc760

Lokasyon Computer Science

Tarih 2019-08

Notlar Singapore Ministry of Education Academic Research Fund Tier 1 under MOE's grant ; UAE University

Örnek Metin The dramatic growth of HTTP adaptive streaming (HAS) traffic represents a practical challenge for service providers in satisfying the demand from their customers. Achieving this in a network where multiple players share the network capacity has so far proved hard because of the bandwidth competition among the HAS players. This competition is exacerbated by the bandwidth overestimation that is introduced due to the isolated and selfish behavior of the HAS players. Each player strives individually to select the maximum bitrate without considering the co-existing players or network resource dynamics. As a result, the HAS players suffer from video quality instability, quality unfairness, and network underutilization or oversubscription, and the players observe a poor quality of experience (QoE). To address this issue, we propose a fully distributed game theory and consensus-based collaborative adaptive bitrate solution for shared network environments, termed Game Theory and consensus-based Approach for Cooperative HAS delivery systems (GTAC). Our solution consists of two-stage games that run in parallel during a streaming session. We extensively evaluate GTAC on a broad set of trace-driven and real-world experiments. Results show that GTAC enhances the viewer QoE by up to 22%, presentation quality stability by up to 24%, fairness by at least 31%, and network utilization by 28% compared to the well-known schemes.

DOI 10.1145/3336496

Cilt 15