| Author: | Xiao, Jie |
| Title: | On evaluating online DASH systems |
| Advisors: | Luo, Xiapu (COMP) |
| Degree: | M.Sc. |
| Year: | 2016 |
| Subject: | Data transmission systems -- Evaluation. Streaming technology (Telecommunications) Hong Kong Polytechnic University -- Dissertations |
| Department: | Department of Computing |
| Pages: | viii, 44 pages : color illustrations |
| Language: | English |
| Abstract: | HTTP streaming has become a cost-effective way for multimedia delivery nowadays. Recently, many DASH (dynamic adaptive streaming over HTTP) algorithms have been proposed to get good performance. As HTML5 video players are adopted by more and more mainstream companies, how these algorithms work on HTML5 players becomes a big concern. In our paper, we come up with an automatically DASH algorithm evaluation system based on the BenchLab framework developed by the University of Massachusetts Amherst. Our system incorporates most of recently proposed DASH algorithms to the system, and we discuss trade-off of these algorithms by carrying out experiments based on the very native HTML5 video player. The result data for comparison are not only in terms of bit rate level and buffer size, but also the quality of experience and fairness. DASH algorithms may be easily implemented and compared when you are willing to develop new algorithms. Using our testbed, we discovered that network fluctuation, number of competing clients, server-side configurations, media segmentations as well as media content itself have influence on the adaptive behavior of algorithms. In general, throughput-based adaption methods perform well in terms of average bit rate of video but have more vulnerabilities of stallings compared to buffer-based algorithms when available bandwidth fluctuates. Buffer-based DASH algorithms have shorter initial delay time and are stable to short-term bandwidth changes without significant degradation of average bit rate level, and they can get better performance for larger buffer size and longer video duration. Fairness algorithms tackle the problem of competing peers sharing available bandwidth unequally, caused by the on-off mode of adaption behavior. However, it pays the cost of degrading video quality. |
| Rights: | All rights reserved |
| Access: | restricted access |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| b28564157.pdf | For All Users (off-campus access for PolyU Staff & Students only) | 1.59 MB | Adobe PDF | View/Open |
Copyright Undertaking
As a bona fide Library user, I declare that:
- I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
- I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
- I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.
By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.
Please use this identifier to cite or link to this item:
https://theses.lib.polyu.edu.hk/handle/200/8380