Improving QoE of Video Streaming using Network Awareness

Over the past few years, there has been a significant increase in the internet traffic due to video content. It is expected that soon 82 percent [2] of global IP traffic will constitute video content. HTTP DASH has been one of the popular on-demand video streaming formats. On the network side, software-defined networking has been a new networking concept which decouple control and data plane to achieve better centralized network control. In this work, we present a cross-layer approach which utilizes the underlying network traffic information to improve the quality of experience of video streaming on SDN networks by selecting least loaded routes. We have evaluated our approach in a geographically distributed OpenFlow testbed using GENI. We also present a model which would utilize the client characteristics to provision quality-of-service. These model have been tested on a number of representative video datasets and various end client adaptation schemes to understand the behaviour at various load scenarios of the network. We observe that our cross-layer approach of selecting least loaded routes in the SDN framework helps in better video streaming experience compared to the case when routes are static. In particular, our specified approach performs better in terms of convergence time and reduces bit rate switching events than traditional DASH streaming.

Name thesis:degree_name

M.S.

Level thesis:degree_level

Masters

Discipline thesis:degree_discipline

Computer Science (UMKC)

Grantor dc:publisher

University of Missouri--Kansas City

Year dc:date.issued

2016

Author dc:creator

• Muppala, Goutham K T

Advisor dc:contributor.advisor

• Medhi, Deepankar

Language dc:language.iso

en_US

Handle dc:identifier.uri

https://hdl.handle.net/10355/60053

OAI identifier oai:identifier

oai:mospace.umsystem.edu:10355/60053