Performance analysis of cell discarding techniques for best effort video communications over ATM networks

Abstract

With increasing interest in the transmission of audio–visual applications over ATM best effort services, efficient video-oriented control mechanisms for improving the video quality in the presence of loss have to be designed. In this paper, we propose and evaluate two new slice-based discard schemes for use with available bit rate and guaranteed frame rate services (e.g. formerly UBR+). The schemes adaptively and selectively adjust the discard level to switch buffer occupancy and video cell payload types. To improve their performance, we also introduce a dynamic frame-level priority data partition technique based on MPEG data structure and feedback from the network. To support these mechanisms, enhancements to the ATM adaptation layer 5 and a new MPEG-2 encapsulation strategy are also proposed. The presented quality of picture (QoP) control framework is evaluated using simulation and actual MPEG video data. The overall aim of the framework is double. First, ensuring a graceful picture quality degradation by minimizing cell loss probability for critical video data, and second optimizing the network effective throughput by reducing transmission of non useful data. In comparison to previous approaches, the performance evaluation have shown a significant reduction of the bad throughput and minimization of losses of intra- and predictive-coded frames at both cell and slice layers.