Efficient transportation of VBR video traffic in ATM networks

The issue of reducing the complexity and yet providing a better Quality of Service (QoS) for effectively supporting real-time traffic in high-speed ATM networks is one of the key research goals of this dissertation work.;The online traffic measurement-based adaptive bandwidth allocation schemes are essential for a quick adaptation to the changing traffic rates and minimizing the number of renegotiations for bandwidth, especially in the context of highly correlated VBR video traffic. The correlation properties of video traffic make predictor-based bandwidth allocation schemes more attractive. An issue that arises from the prediction of video traffic is the effect of error in traffic estimation on the queues. These issues are addressed in the proposed video traffic management architecture.;A novel predictor-based architecture for traffic management of VBR video traffic and the associated algorithms are proposed in this dissertation. The algorithms incorporate the dynamic bandwidth allocation at burst level (milliseconds) and a short-term resource management through cell scheduling (microseconds). Linear predictors (regressive) of VBR video traffic were employed in a session-wise Predictor-based Dynamic Bandwidth Allocation (PDBA) scheme with low computational requirements. A unique feature of this architecture is the Short-Term Controller (STC) that minimizes the ill-effects of traffic prediction errors on the queuing system.;In this dissertation work, investigations are made to improve the utilization factor for a desired cell loss probability (CLP) based on online traffic estimation. A simple Least Mean Square (LMS) error based prediction system has been employed along with buffer monitoring. A buffer threshold (BT) parameter is introduced and a backlog (buffered cells) beyond the BT, denoted as unsafe state, triggers a higher band-width reallocation (based on the maximum of the predictor estimates and the current allocation values). In the safe state (backlog less than BT), the system switches back to the bandwidth of a target utilization. Properties of the system are studied and conditions under which cell loss can be minimized are derived. (Abstract shortened by UMI.).