Modeling and analysis of digital video transmission over networks using packet-level FEC

Forward error correction (FEC) has often been proposed to recover packet losses and improve packet video transport in packet-switched communication networks. The author first study the efficacy of FEC coding at the network layer in terms of the effective packet-loss rate. The approach considers a simplified network scenario described in terms of a single bottleneck node, modeled by a single multiplexer, and then finds the optimum compromise between the packet-loss recovery capability of FEC and the increased raw packet-loss rate caused by the overhead associated with transmitting redundant packets. The analysis shows that, contrary to some previous conclusions, FEC has significant potential in reducing residual packet-loss rate provided that the code rate is chosen appropriately. By modeling the fully-interleaved network transport channel as a block interference channel (BIC), an information-theoretic bound on the performance achievable with FEC is provided. We then proceed to investigate the efficacy of FEC from the perspective of the video application layer and assess the effectiveness of FEC coding using the end-to-end video quality as the evaluation metric. A model-based framework is developed to describe the FEC-protected packet video network transport system which is then used to analytically investigate the overall efficacy of packet-level FEC in improving the end-to-end video quality. We provide an analysis of the delay caused by FEC coding/decoding and analytically investigate the effects of several important system parameters on the effectiveness of FEC and the optimal selection of FEC coding parameters subject to these constraints. Another frequently used technique for improving transport performance on lossy packet networks is multiple description coding (MDC). Typically MDC and FEC are used separately and independently with resulting performance dependent on network conditions, the choice of coding parameters and the applications being transported. In this dissertation we propose and analyze a joint MDC and FEC coding approach for delay-constrained applications on congested networks under some idealized modeling assumptions which are, nevertheless, sufficient to draw some qualitative conclusions. Path diversity has recently been proposed to improve network transport for both single-description (SD) and multiple-description (MD) coded video. In this work we model and analyze an SD coded video transmission system employing packet-level FEC in combination with path diversity. In particular, we provide a precise analytical approach to evaluating the efficacy of path diversity in reducing the burstiness of network packet-loss processes. We use this approach to quantitatively demonstrate the advantages of path diversity in improving end-to-end video transport performance using packet-level FEC. As part of this work we also investigate the accuracy of Gilbert models and other low-complexity discrete-time Markov chain models in characterizing the packet-loss process associated with a transport network whose behavior can be described in terms of a single bottleneck node, modeled by a single multiplexer. It is demonstrated that, although higher-order Markov chain models can achieve increasingly more accurate descriptions, the Gilbert model has some serious deficiencies in predicting the packet-loss statistics of the single-multiplexer model for a variety of packet arrival processes. We show that this has some serious consequences for the performance evaluation of forward error correction (FEC) coding schemes using Markov chain models compared to that predicted by an exact queueing analysis of the single-multiplexer model