This paper addresses the transport of real-time multimedia traffic generated by MPEG-2 applications over ATM networks using an enhanced UBR best effort service (UBR+). Based on the factors affecting the picture quality during transmission, we propose an efficient and cost-effective ATM best effort delivery service. The proposed service integrates three components: a dynamic frame level priority assignation mechanism based on the MPEG data structure and feedback from the network (DexPAS), a novel audiovisual AAL5 SSCS with FEC, and an intelligent packet video discard scheme named SA-PSD, which adaptively and selectively adjusts the cell drop level to switch buffer occupancy, video cell payload type and forward error correction ability of the destination. The overall best effort video delivery framework is evaluated using ATM network simulation and MPEG-2 video traces. The ultimate aim of this framework is twofold. First, minimizing loss for critical video data with bounded end-to-end delay for arriving cells. Second, reducing the bad throughput crossing the network during congestion. Compared to previous approaches, performance evaluation shows a good protection of predictive coded and bidirectional predictive coded frames at the video slice layer.

AAL1 represents a long term solution for the transport of CBR MPEG2 video traffic over ATM networks. This paper describes an integrated implementation model for AAL1 carrying CBR MPEG2 traffic with particular focus on the receiver convergence sublayer functions and the issues affecting the end-to-end performance of the AAL1 connection. Through a comprehensive simulation of the AAL and the ATM layers, we verify that the implementation model is capable of delivering the desired QoS for video applications under a wide range of cell losses and cell delay variations. In addition, we show the effects of using FEC with byte interleaving on the end-to-end performance parameters. We conclude that cell loss statistics constitute the key factors in determining the overall performance of the AAL1 connection.