DocumentCode
2145629
Title
A Novel SVC VoD System with Rate Adaptation and Error Concealment over GPRS/EDGE Network
Author
Zhang, Li-Na ; Yuan, Chun ; Zhong, Yu-Zhuo
Author_Institution
Dept. of Comput. Sci. & Technol., Tsinghua Univ., Beijing
Volume
1
fYear
2008
fDate
27-30 May 2008
Firstpage
349
Lastpage
354
Abstract
Considering the high packet losses and low, varying bandwidth of GPRS/EDGE network and limited computation power of handheld devices, we present and implement a novel SVC video-on-demand system for hand-held devices over GPRS/EDGE network. For the purpose of handling varying bitrate, we propose a priority based layer switching (PLS) adaptation scheme for SVC stream, which not only performs online 3-D adaptation more quickly by a simple parser, but also optimizes the video quality in a R-D sense under the bandwidth constraint. To resist packet losses, a motion-detection based adaptive error concealment (MDA) algorithm is proposed, which can achieve a PSNR gain of up to 3dB comparing to existing methods, while maintaining low complexity. Moreover, our proposed system was implemented and tested over existing GPRS/EDGE network deployed in China. The test results demonstrate that the proposed system and schemes have performance advantage in terms of quicker data rate adaptation, higher PSNR and lower overhead.
Keywords
3G mobile communication; packet radio networks; telecommunication switching; China; GPRS/EDGE network; SVC VoD System; bandwidth constraint; handheld devices; motion-detection based adaptive error concealment; priority based layer switching adaptation scheme; rate adaptation; video quality; video-on-demand system; Bandwidth; Bit rate; Computer networks; Constraint optimization; Ground penetrating radar; Handheld computers; PSNR; Static VAr compensators; Streaming media; System testing; H.264/AVC SVC; SVC error concealment; VoD; mobile; rate adaptation;
fLanguage
English
Publisher
ieee
Conference_Titel
Image and Signal Processing, 2008. CISP '08. Congress on
Conference_Location
Sanya, Hainan
Print_ISBN
978-0-7695-3119-9
Type
conf
DOI
10.1109/CISP.2008.315
Filename
4566177
Link To Document