QoS Control Strategies for High-Quality Video Processing

Authors:
Clemens C. Wust;Liesbeth Steffens;Reinder J. Bril;Wim F. J. Verhaegh
Affiliations:
Philips Research Laboratories;Philips Research Laboratories;Philips Research Laboratories and Eindhoven University of Technology;Philips Research Laboratories
Venue:
ECRTS '04 Proceedings of the 16th Euromicro Conference on Real-Time Systems
Year:
2004

Citing 0
Cited 8

Fine Grain QoS Control for Multimedia Application Software

Proceedings of the conference on Design, Automation and Test in Europe - Volume 2
Dynamic QoS-Aware Coalition Formation

IPDPS '05 Proceedings of the 19th IEEE International Parallel and Distributed Processing Symposium (IPDPS'05) - Workshop 2 - Volume 03
QoS control for optimality and safety

Proceedings of the 5th ACM international conference on Embedded software
Hierarchical resource allocation for robust in-home video streaming

Journal of Systems and Software
Symbolic quality control for multimedia applications

Real-Time Systems
Dynamic tuning of feature set in highly variant interactive applications

EMSOFT '10 Proceedings of the tenth ACM international conference on Embedded software
An ada framework for QoS-aware applications

Ada-Europe'05 Proceedings of the 10th Ada-Europe international conference on Reliable Software Technologies
A framework for video streaming to resource-constrained terminals

EUC'05 Proceedings of the 2005 international conference on Embedded and Ubiquitous Computing

Quantified Score

Hi-index	0.00

Visualization

Abstract

Video processing in software is often characterized byhighly fluctuating, content-dependent processing times, anda limited tolerance for deadline misses. We present anapproach that allows close-to-average-case resource allocationto a single video processing task, based on asynchronous,scalable processing, and QoS adaptation. TheQoS adaptation balances different QoS parameters that canbe tuned by user-perception experiments: picture quality,deadline misses, and quality changes. We model the balancingproblem as a discrete stochastic decision problem, andpropose two closely related solution strategies, for whichthe processing-time statistics are determined off line and atrun time, respectively. We enhance both strategies with acompensation for structural (non-stochastic) load fluctuations.Finally, we validate our approach by means of simulationexperiments, and conclude that both enhanced strategiesperform close to the theoretical optimum.