Autonomic Resource Management for Multimedia Services Using Inventory Control
MMNS '07 Proceedings of the 10th IFIP/IEEE International Conference on Management of Multimedia and Mobile Networks and Services: Real-Time Mobile Multimedia Services
Operations Research Methods for Autonomic Resource Management
Journal of Network and Systems Management
| Hi-index | 0.00 |
Flow and admission control algorithms have been proposed for efficient resource utilization in wireless networks with scarce and randomly varying resource availability. Transport and lower layer control protocols administer generic control behavior, ignoring particular application requirements or characteristics. Network-aware applications, on the other hand, enable us to develop application specific control mechanisms. In this paper, we have proposed a novel control theory based adaptive resource management framework for network-aware scalable multimedia applications in wireless networks. In particular, we design an application-aware pro-active middleware based on level crossing analysis of stochastic models assisting a set of control algorithms founded on automatic control theory, thus avoiding network congestion and ensuring optimal resource utilization and application level QoS. Simulation experiments show that the resource management framework is capable of achieving bandwidth utilization within 1.5% of the optimum (theoretical) value and a delay jitter of 1.3 msec虏 only, for real-time video streaming. Also, it ensures a low session blocking rate (less than 10% for high session arrival rate) and high percentage of sessions with good quality video.