MixNStream: multi-source video distribution with stream mixers

Authors:
Philip Yuen;Gary Chan
Affiliations:
The Hong Kong University of Science and Technology, Kowloon, Hong Kong;The Hong Kong University of Science and Technology, Kowloon, Hong Kong
Venue:
Proceedings of the 2010 ACM workshop on Advanced video streaming techniques for peer-to-peer networks and social networking
Year:
2010

Citing 6
Cited 0

A distributed algorithm for constructing minimum delay spanning trees under bandwidth constraints on overlay networks

Systems and Computers in Japan
Additive guarantees for degree bounded directed network design

STOC '08 Proceedings of the fortieth annual ACM symposium on Theory of computing
Distributed Algorithms for Constructing Approximate Minimum Spanning Trees in Wireless Sensor Networks

IEEE Transactions on Parallel and Distributed Systems
OMNI: An efficient overlay multicast infrastructure for real-time applications

Computer Networks: The International Journal of Computer and Telecommunications Networking - Overlay distribution structures and their applications
Low-delay mesh with peer churns for peer-to-peer streaming

ICME'09 Proceedings of the 2009 IEEE international conference on Multimedia and Expo
Fast-mesh: a low-delay high-bandwidth mesh for peer-to-peer live streaming

IEEE Transactions on Multimedia

Quantified Score

Hi-index	0.00

Visualization

Abstract

Many Internet streaming applications, such as distributed surveillance, multimedia webcasting and video thumbnails for Internet TV channel browsing, require video mixers where streams are aggregated as a single stream before presenting to users. We consider in this paper a multi-source streaming network with distributed mixers, where streams originated from multiple sources are mixed before presented to distributed users. We are interested in minimizing the worst-case delay from the source to users via the mixers. We propose an adaptive and distributed protocol called MixNStream, which continuously reduces the network diameter in the presence of churns. Through simulations on Internet-like topologies, we show that MixNStream achieves substantially better performance as compared with the state-of-the-art in terms of network delay and network stress.