A hybrid error recovery scheme for scalable video transmission over 3G cellular broadcast networks
- Authors
- Kang, Kyungtae; Cho, Yongwoo; Shin, Heonshik
- Issue Date
- Feb-2009
- Publisher
- Baltzer Science Publishers B.V.
- Keywords
- cdma2000 1xEV-DO; Reed-Solomon; Hybrid error recovery; BCMCS; MPEG-4 FGS
- Citation
- Wireless Networks, v.15, no.2, pp 241 - 258
- Pages
- 18
- Indexed
- SCIE
SCOPUS
- Journal Title
- Wireless Networks
- Volume
- 15
- Number
- 2
- Start Page
- 241
- End Page
- 258
- URI
- https://scholarworks.bwise.kr/erica/handle/2021.sw.erica/41412
- DOI
- 10.1007/s11276-007-0038-5
- ISSN
- 1022-0038
1572-8196
- Abstract
- The cdma2000 1xEV-DO mobile communication system provides broadcast and multicast services (BCMCS) to meet an increasing demand for multimedia data services. But the servicing of video streams over a BCMCS network faces a challenge from the unreliable and error-prone nature of the radio channel. BCMCS uses Reed-Solomon coding integrated with the MAC protocol for error recovery. We show that this is not effective for mobiles moving at the edge of service area, where the channel condition is bad, resulting in significantly lower video quality. To improve the playback quality of an MPEG-4 FGS (fine granularity scalability) video stream, we propose a hybrid error recovery scheme incorporating a packet scheduler, which uses slots saved by reducing the Reed-Solomon coding overhead. Packets to be retransmitted are prioritized by a utility function which reduces the packet error-rate in the application layer within a fixed retransmission budget by considering of the map of the error control block at each mobile node. Our error recovery scheme also uses the characteristics of MPEG-4 FGS to improve the video quality even for a slow-moving mobile which is experiencing a high error-rate in the physical channel because of error bursts.
- Files in This Item
-
Go to Link
- Appears in
Collections - COLLEGE OF COMPUTING > SCHOOL OF MEDIA, CULTURE, AND DESIGN TECHNOLOGY > 1. Journal Articles
- COLLEGE OF COMPUTING > DEPARTMENT OF ARTIFICIAL INTELLIGENCE > 1. Journal Articles

Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.