A Comparative Study of DASH Representation Sets Using Real User Characteristics

Back to the main website of this dataset

If you use any part of this dataset for your research, we kindly ask you to refer this paper in any publication mentioning or using our dataset:

Christian Kreuzberger, Benjamin Rainer, Hermann Hellwagner, Laura Toni, Pascal Frossard: "A Comparative Study of DASH Representation Sets Using Real User Characteristics", submitted to NOSSDAV 2016.

User Base

Original Dataset

We generated our User Base based on YouTube Live Traces from K. Pires, G. Simon: YouTube live and Twitch: a tour of user-generated live streaming systems (Link to paper, Link to their website).

Generated Scenarios

We created ready-to-go request files in CSV format, which are available for download here:

• 2014-1-6
• 2014-1-7
• 2014-1-8
• 2014-1-9
• 2014-1-10
• 2014-1-11
• 2014-1-12
• 2014-1-13
• 2014-1-14
• 2014-1-15
• 2014-1-16

Figures

By Video Category

The figures below show the distribution over time per video category and per screen resolution. Note that the figures represent a stacked plot, showing the distribution between categories/screen resolutions, and not the total numbers.
The figures below show the distribution over time per video category.

2014-1-6

2014-1-7

2014-1-8

2014-1-9

2014-1-10

2014-1-11

2014-1-12

2014-1-13

2014-1-14

2014-1-15

2014-1-16