FEC-based multiple description coding for heterogeneous client bandwidths

We consider applying the FEC-based multiple description coding method for multicasting a source to a set of clients with heterogeneous access bandwidths. In the scenario where different clients access the server via separate links, we propose a procedure which refines the already computed optimal solutions for other bandwidths, using different packet lengths. Compared to the scheme that computes an optimal solution individually for each bandwidth, this procedure is quite faster and has comparable performance. In another scenario where many clients share a bottleneck link, we extend our embedded packetization framework to be available in more than two bandwidths, and propose a fast heuristic algorithm which can achieve very good performance tradeoff among all clients.