Adaptive buffer control to minimise delay and guarantee service reliability
- Authors
- Kang, K.; Suh, H. -J.
- Issue Date
- Jan-2011
- Publisher
- INST ENGINEERING TECHNOLOGY-IET
- Keywords
- REED-SOLOMON CODES; PERFORMANCE; BROADCAST; SYSTEMS; CHANNELS
- Citation
- IET COMMUNICATIONS, v.5, no.1, pp 110 - 118
- Pages
- 9
- Indexed
- SCI
SCIE
SCOPUS
- Journal Title
- IET COMMUNICATIONS
- Volume
- 5
- Number
- 1
- Start Page
- 110
- End Page
- 118
- URI
- https://scholarworks.bwise.kr/erica/handle/2021.sw.erica/38307
- DOI
- 10.1049/iet-com.2010.0070
- ISSN
- 1751-8628
1751-8636
- Abstract
- Delay-sensitive real-time applications require average latency to be reduced as far as possible, while maintaining the minimum necessary level of service quality. However, the interaction of these two performance metrics has not received much attention in the current standard air specification of CDMA2000 broadcast services. The authors analyse the performance of the forward error correction employed at the MAC layer of these broadcast services, with respect to the conditions of a fading wireless channel and the size of the error control block (ECB), which is a buffer in the form of a two-dimensional array. As a result, the authors establish the tradeoff between service reliability and delay, and show how it depends on the size of the ECB. The authors go on to propose a novel adaptive error-recovery scheme which dynamically adjusts the size of the ECB to suit the wireless environment of the mobile nodes, using either a moving-average or a weighted-average method. Reducing the service delay as far as possible necessarily compromises the reliability of the service, but the authors scheme still ensures the required minimum reliability. The authors have shown the effectiveness of their scheme through extensive simulations using a detailed threshold model, which accurately describes the statistics of the block error process in a fading channel. The authors also suggest a service implementation scenario to aid in the design.
- Files in This Item
-
Go to Link
- Appears in
Collections - COLLEGE OF COMPUTING > DEPARTMENT OF ARTIFICIAL INTELLIGENCE > 1. Journal Articles

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