Coding and Scheduling for Efficient Loss-Resilient Data Broadcasting

Abstract

We examine the problem of sending data to clients over a broadcast channel in a way that minimizes the expected waiting time of the clients for this data. This channel, however, is not completely reliable, and packets are occasionally lost. This poses a problem, as performance is greatly degraded by even a single packet loss. For example, one lost packet will increase our expected waiting time for an item from .75 to 2 or 167%, when sending two items with equal demands. We propose and analyze two solutions that attempt to minimize this degradation. In the first, we code packets and in the second we code packets and slightly modify our schedule. The resulting degradations are 67% for the first solution and less than 1% for the second. We conclude that using the second scheme is a very effective way to combat single packet losses, and we extend this solution to combat up to t packet losses per data item for any t [ ] k, where k is the number of packets per data item.