Split-layer video multicast protocol: a new receiver-based rate-adaptation protocol

Abstract

This paper proposes a new receiver-based rate-adaptation protocol for multicasting video, called Split-Layer Video Multicast (SPLIT). Unlike existing receiver-based rate-adaptation protocols, such as Receiver-driven Layered Multicast (RLM), the SPLIT protocol is specifically designed to take advantage of existing packet loss concealment techniques to provide end-users with increased quality of video. In an effort to gauge the performance of the SPLIT protocol, a number of experiments using the NS-2 network simulator were conducted and the results were compared with those of the RLM protocol. The results show that the proposed SPLIT protocol utilizes the available bandwidth in a more efficient way than the RLM protocol does.