Design of a Scalable Proxy for Peer-to-Peer Video Streaming Using Multiple Description Coding

• Main Authors: Shi-Yuan Lan, 籃士淵
• Other Authors: Jen-Wen Ding
• Format: Others
• Language: zh-TW
• Published: 2007
• Online Access: http://ndltd.ncl.edu.tw/handle/06946025270699336380

碩士 === 國立高雄應用科技大學 === 資訊管理研究所碩士班 === 95 === With rapid development of network technologies and multimedia applications, there is an increasing demand for large-scale on-demand video streaming. However, due to the large volume of video data, traditional client-server architecture is incapable of providing large-scale video-on-demand streaming. In recent years, several solutions have been proposed to tackle this difficulty. These solutions can be classified into two categories: infrastructure-based and infrastructure-less. The infrastructure-based approach is known as content delivery network (CDN) approach, which deploys numerous extra CDN servers (or proxies) at boundary networks. The main drawback of the CDN approach is that it is costly to deploy the numerous CDN servers. The infrastructure-less approach is peer-to-peer (P2P) streaming. The main drawback of the P2P approach is its instability due to the frequent online/offline behavior of peers. In view of this, in this paper, we combine the two types of approaches. We develop a P2P-based video proxy by incorporating multiple description coding. We call our method MPPP (MDC-based Peer-to-Peer Proxy). The main merits of MPPP is that it is high adaptive to heterogeneous network environments, so it can help heterogeneous clients to obtain different video qualities. Our simulation results show that MPPP is superior to the conventional P2P-based video proxy in many ways, especially in heterogeneous network environment, such as reducing the network traffic, decreasing the server loading, and improving clients’ perceptual video quality.