Layer-Aware FEC Based Scalable Multiple Description Coding for Robust Video Transmission over Path Diversity Networks

Dinh Trieu DUONG; Deepak Kumar SINGH; Seok Ho WON; Doug Young SUH

doi:10.1587/transcom.E96.B.2323

Layer-Aware FEC Based Scalable Multiple Description Coding for Robust Video Transmission over Path Diversity Networks

Dinh Trieu DUONG, Deepak Kumar SINGH, Seok Ho WON, Doug Young SUH

Full Text Views

0

Cite this

Summary :

In this paper, we propose a novel layered scalable- multiple description coding (LS-MDC) which offers the benefits of both scalable video coding and multiple description coding for robust video transmission over packet lossy networks. In the proposed LS-MDC method, multiple descriptions including base layer, enhancement layers, and their corresponding FEC parity data are allocated into two network paths of a path diversity system. Unlike the conventional approaches, the source base/enhancement data and their own parities in the proposed method are not transmitted together but are transferred over different paths. Therefore, the effect of burst packet losses can be effectively reduced for the descriptions. Furthermore, in order to minimize the overall distortion for the LS-MDC system and exploit the benefits of path diversity, we also propose an optimal rate allocation scheme that can adaptively control the transmission rate as well as the channel coding rate for media senders. Experiments show that the proposed method provides much better peak signal-to-noise ratio (PSNR) than conventional MDC techniques.

Publication: IEICE TRANSACTIONS on Communications Vol.E96-B No.9 pp.2323-2332

Publication Date: 2013/09/01

Publicized

Online ISSN: 1745-1345

DOI: 10.1587/transcom.E96.B.2323

Type of Manuscript: PAPER

Category: Multimedia Systems for Communications

Authors

Dinh Trieu DUONG
  Kyunghee University
Deepak Kumar SINGH
  Kyunghee University
Seok Ho WON
  Electronics & Telecommunication Research Institute (ETRI)
Doug Young SUH
  Kyunghee University

Keyword

multiple description coding (MDC), scalable video coding (SVC), forward error correction (FEC), path diversity

Cite this

Copy

Dinh Trieu DUONG, Deepak Kumar SINGH, Seok Ho WON, Doug Young SUH, "Layer-Aware FEC Based Scalable Multiple Description Coding for Robust Video Transmission over Path Diversity Networks" in IEICE TRANSACTIONS on Communications, vol. E96-B, no. 9, pp. 2323-2332, September 2013, doi: 10.1587/transcom.E96.B.2323.
Abstract: In this paper, we propose a novel layered scalable- multiple description coding (LS-MDC) which offers the benefits of both scalable video coding and multiple description coding for robust video transmission over packet lossy networks. In the proposed LS-MDC method, multiple descriptions including base layer, enhancement layers, and their corresponding FEC parity data are allocated into two network paths of a path diversity system. Unlike the conventional approaches, the source base/enhancement data and their own parities in the proposed method are not transmitted together but are transferred over different paths. Therefore, the effect of burst packet losses can be effectively reduced for the descriptions. Furthermore, in order to minimize the overall distortion for the LS-MDC system and exploit the benefits of path diversity, we also propose an optimal rate allocation scheme that can adaptively control the transmission rate as well as the channel coding rate for media senders. Experiments show that the proposed method provides much better peak signal-to-noise ratio (PSNR) than conventional MDC techniques.
URL: https://global.ieice.org/en_transactions/communications/10.1587/transcom.E96.B.2323/_p

Copy

@ARTICLE{e96-b_9_2323,
author={Dinh Trieu DUONG, Deepak Kumar SINGH, Seok Ho WON, Doug Young SUH, },
journal={IEICE TRANSACTIONS on Communications},
title={Layer-Aware FEC Based Scalable Multiple Description Coding for Robust Video Transmission over Path Diversity Networks},
year={2013},
volume={E96-B},
number={9},
pages={2323-2332},
abstract={In this paper, we propose a novel layered scalable- multiple description coding (LS-MDC) which offers the benefits of both scalable video coding and multiple description coding for robust video transmission over packet lossy networks. In the proposed LS-MDC method, multiple descriptions including base layer, enhancement layers, and their corresponding FEC parity data are allocated into two network paths of a path diversity system. Unlike the conventional approaches, the source base/enhancement data and their own parities in the proposed method are not transmitted together but are transferred over different paths. Therefore, the effect of burst packet losses can be effectively reduced for the descriptions. Furthermore, in order to minimize the overall distortion for the LS-MDC system and exploit the benefits of path diversity, we also propose an optimal rate allocation scheme that can adaptively control the transmission rate as well as the channel coding rate for media senders. Experiments show that the proposed method provides much better peak signal-to-noise ratio (PSNR) than conventional MDC techniques.},
keywords={},
doi={10.1587/transcom.E96.B.2323},
ISSN={1745-1345},
month={September},}

Copy

TY - JOUR
TI - Layer-Aware FEC Based Scalable Multiple Description Coding for Robust Video Transmission over Path Diversity Networks
T2 - IEICE TRANSACTIONS on Communications
SP - 2323
EP - 2332
AU - Dinh Trieu DUONG
AU - Deepak Kumar SINGH
AU - Seok Ho WON
AU - Doug Young SUH
PY - 2013
DO - 10.1587/transcom.E96.B.2323
JO - IEICE TRANSACTIONS on Communications
SN - 1745-1345
VL - E96-B
IS - 9
JA - IEICE TRANSACTIONS on Communications
Y1 - September 2013
AB - In this paper, we propose a novel layered scalable- multiple description coding (LS-MDC) which offers the benefits of both scalable video coding and multiple description coding for robust video transmission over packet lossy networks. In the proposed LS-MDC method, multiple descriptions including base layer, enhancement layers, and their corresponding FEC parity data are allocated into two network paths of a path diversity system. Unlike the conventional approaches, the source base/enhancement data and their own parities in the proposed method are not transmitted together but are transferred over different paths. Therefore, the effect of burst packet losses can be effectively reduced for the descriptions. Furthermore, in order to minimize the overall distortion for the LS-MDC system and exploit the benefits of path diversity, we also propose an optimal rate allocation scheme that can adaptively control the transmission rate as well as the channel coding rate for media senders. Experiments show that the proposed method provides much better peak signal-to-noise ratio (PSNR) than conventional MDC techniques.
ER -