Prefetching Scheme in Streaming System Based on Interactive Media Access Patterns

Abstract:

User behavior modeling is critical to help reduce delays due to buffering before and during a video playback at a client. In this paper, we model the states a user transitions while browsing through videos to be the hidden states of a Hidden Markov Model. The users next seeking position was estimated according to the maximum posteriori criterion. The proposed prefetching schemes chose and download the data block which had the higher caching value in advance to reduce data access delay. We evaluate and compare the proposed schemes using user browsing pattern data and find that prefetching scheme can improve the seeking performance significantly.

• Periodical: Applied Mechanics and Materials (Volumes 411-414)
• Pages: 966-969
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.411-414.966
• Online since: September 2013
• Authors: Xi Wen Zhang*, Gang Wu, Song Wang
• Keywords: Caching Policy, Hidden Markov Model (HMM), Quality of Service (QoS), User Behavior Analysis

References

[1] Y. HE, G. Shen and Y. Xiong: IEEE Transactions on Multimedia, Vol. 11 (2009) p.138.

[2] S. Khemmarat and R. Zhou: Signal Processing: Image Communication, Vol. 27 (2012), p.343.

[3] J. Choi and A.S. Reaz: IEEE Communications Surveys and Tutorials, Vol. 14 (2012), p.156.

[4] L. GUO, E. Tan and S. Chen, in Proc. the Annual ACM Symposium on Principles of Distributed Computing (2009), p.283.

[5] W. Zheng, G. Shen and S. Li: Int. J. Digital Content Technology and its Applications, Vol. 6 (2012), p.153.

[6] A. Mohamed, S. Spagna and F. Huici, in Pro. the 6th ACM International Conference on Web Search and Data Mining, (2013), p.607.

[7] C. Zhen, G. Shen and S. Li, in Proc. the ACM workshop on Advances in peer-to-peer multimedia streaming (2012), p.29.