Feedback-based adaptive video streaming over lossy channels

Jong-Ok Kim, Hideki Tode, Koso Murakami

Research output: Contribution to journalArticle

Abstract

In packet networks including the Internet and commercial 3G wireless bearers, the network states that a streaming media application experiences are not known a priori and exhibit time-varying characteristics. For such dynamic environments, network-adaptive techniques are essential to efficiently deliver video data. In this paper, we propose a frame-based optimal scheduling algorithm which incorporates a MAP (Maximum A Posteriori) framework for adapting to varying network loss rate. The optimal transmission schedule is determined such that effective frame-rate is maximized at playback. Also, for multiple packets per frame, frame-based selection of delivery order greatly reduces computational complexity for a server scheduler when compared with packet-based scheduling techniques. In addition, by dynamically estimating instantaneous packet loss probability, the proposed scheduler performs network-adaptive transmission for streaming video over time-varying packet networks. Simulation results for test video sequence show that the proposed scheduling algorithm outperforms conventional ARQ-based schemes from a view point of reconstructed video quality as well as playable frame-rate. In particular, the proposed scheduling algorithm exhibits significant improvements of frame-rate over highly lossy channels.

Original languageEnglish
Pages (from-to)3076-3084
Number of pages9
JournalIEICE Transactions on Communications
VolumeE87-B
Issue number10
Publication statusPublished - 2004 Oct 1
Externally publishedYes

Fingerprint

Video streaming
Scheduling algorithms
Packet networks
Feedback
Time varying networks
Media streaming
Packet loss
Computational complexity
Servers
Scheduling
Internet

Keywords

  • ARQ
  • Lossy channel
  • Packet loss recovery
  • Packet scheduling
  • Streaming video

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Computer Networks and Communications

Cite this

Feedback-based adaptive video streaming over lossy channels. / Kim, Jong-Ok; Tode, Hideki; Murakami, Koso.

In: IEICE Transactions on Communications, Vol. E87-B, No. 10, 01.10.2004, p. 3076-3084.

Research output: Contribution to journalArticle

Kim, J-O, Tode, H & Murakami, K 2004, 'Feedback-based adaptive video streaming over lossy channels', IEICE Transactions on Communications, vol. E87-B, no. 10, pp. 3076-3084.
Kim, Jong-Ok ; Tode, Hideki ; Murakami, Koso. / Feedback-based adaptive video streaming over lossy channels. In: IEICE Transactions on Communications. 2004 ; Vol. E87-B, No. 10. pp. 3076-3084.
@article{a0dd57b577424b869a9e3b0047fbb515,
title = "Feedback-based adaptive video streaming over lossy channels",
abstract = "In packet networks including the Internet and commercial 3G wireless bearers, the network states that a streaming media application experiences are not known a priori and exhibit time-varying characteristics. For such dynamic environments, network-adaptive techniques are essential to efficiently deliver video data. In this paper, we propose a frame-based optimal scheduling algorithm which incorporates a MAP (Maximum A Posteriori) framework for adapting to varying network loss rate. The optimal transmission schedule is determined such that effective frame-rate is maximized at playback. Also, for multiple packets per frame, frame-based selection of delivery order greatly reduces computational complexity for a server scheduler when compared with packet-based scheduling techniques. In addition, by dynamically estimating instantaneous packet loss probability, the proposed scheduler performs network-adaptive transmission for streaming video over time-varying packet networks. Simulation results for test video sequence show that the proposed scheduling algorithm outperforms conventional ARQ-based schemes from a view point of reconstructed video quality as well as playable frame-rate. In particular, the proposed scheduling algorithm exhibits significant improvements of frame-rate over highly lossy channels.",
keywords = "ARQ, Lossy channel, Packet loss recovery, Packet scheduling, Streaming video",
author = "Jong-Ok Kim and Hideki Tode and Koso Murakami",
year = "2004",
month = "10",
day = "1",
language = "English",
volume = "E87-B",
pages = "3076--3084",
journal = "IEICE Transactions on Communications",
issn = "0916-8516",
publisher = "Maruzen Co., Ltd/Maruzen Kabushikikaisha",
number = "10",

}

TY - JOUR

T1 - Feedback-based adaptive video streaming over lossy channels

AU - Kim, Jong-Ok

AU - Tode, Hideki

AU - Murakami, Koso

PY - 2004/10/1

Y1 - 2004/10/1

N2 - In packet networks including the Internet and commercial 3G wireless bearers, the network states that a streaming media application experiences are not known a priori and exhibit time-varying characteristics. For such dynamic environments, network-adaptive techniques are essential to efficiently deliver video data. In this paper, we propose a frame-based optimal scheduling algorithm which incorporates a MAP (Maximum A Posteriori) framework for adapting to varying network loss rate. The optimal transmission schedule is determined such that effective frame-rate is maximized at playback. Also, for multiple packets per frame, frame-based selection of delivery order greatly reduces computational complexity for a server scheduler when compared with packet-based scheduling techniques. In addition, by dynamically estimating instantaneous packet loss probability, the proposed scheduler performs network-adaptive transmission for streaming video over time-varying packet networks. Simulation results for test video sequence show that the proposed scheduling algorithm outperforms conventional ARQ-based schemes from a view point of reconstructed video quality as well as playable frame-rate. In particular, the proposed scheduling algorithm exhibits significant improvements of frame-rate over highly lossy channels.

AB - In packet networks including the Internet and commercial 3G wireless bearers, the network states that a streaming media application experiences are not known a priori and exhibit time-varying characteristics. For such dynamic environments, network-adaptive techniques are essential to efficiently deliver video data. In this paper, we propose a frame-based optimal scheduling algorithm which incorporates a MAP (Maximum A Posteriori) framework for adapting to varying network loss rate. The optimal transmission schedule is determined such that effective frame-rate is maximized at playback. Also, for multiple packets per frame, frame-based selection of delivery order greatly reduces computational complexity for a server scheduler when compared with packet-based scheduling techniques. In addition, by dynamically estimating instantaneous packet loss probability, the proposed scheduler performs network-adaptive transmission for streaming video over time-varying packet networks. Simulation results for test video sequence show that the proposed scheduling algorithm outperforms conventional ARQ-based schemes from a view point of reconstructed video quality as well as playable frame-rate. In particular, the proposed scheduling algorithm exhibits significant improvements of frame-rate over highly lossy channels.

KW - ARQ

KW - Lossy channel

KW - Packet loss recovery

KW - Packet scheduling

KW - Streaming video

UR - http://www.scopus.com/inward/record.url?scp=7544245628&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=7544245628&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:7544245628

VL - E87-B

SP - 3076

EP - 3084

JO - IEICE Transactions on Communications

JF - IEICE Transactions on Communications

SN - 0916-8516

IS - 10

ER -