On The Effects of Path Correlation in Multi-Path Video Communications using FEC over Lossy.pdf

Abstract—In this paper, we investigate the effect of path correlation on video communications when path-diversity routing techniques are employed together with forward error correction (FEC). We study the statistical properties of packet losses when correlation between paths exists and demonstrate that the effects of path correlation on received video quality depend on various factors, such as the effectiveness of error control strategies (passive error concealment, FEC, etc.) and the existing channel conditions. We demonstrate that, contrary to common belief, path correlation need not be detrimental to the system performance. The results are expected to have some impact on the development of explicit multi-path routing strategies. Keywords—Path correlation, Joint links, Path diversity, FEC, Channel coding, Error concealment, Video quality. I. INTRODUCTION Diversity techniques have been used at the physical layer in wireless communications for many years as a means of improving end-to-end communication performance. Recently, path-diversity-based routing schemes for video transmission have been proposed for packet networks as well [1, 3, 5, 9]. The use of diversity in this case is partly motivated by the observation that if the loss patterns on different paths are independent, then large burst losses are less likely to occur. However, the assumption of path independence is rarely warranted on the Internet. For example, two different Internet paths might share one or more joint links somewhere in the network, which may introduce path correlation between their respective loss processes. To avoid this, methods have been proposed to choose paths with a minimal number of joint links [2, 5]. Most of the proposed schemes (with a notable exception in [4]) assume that correlation among packet losses on different paths (spatial correlation) will degrade system performance. Certainly, this may be true for some specific choices of source coding