## Introduction

In this page, you can download the datasets and matlab source code which are used in our paper: **Kun HE, Gwendal Simon and Patrick Maillé. Watermarked Video Delivery: Traffic Reduction and CDN Management. In Proceedings of ACM Multimedia System (MMSys) 2018**

If you want to download the files, please provide a reference.

## Datasets

Database: output.txt (139.6 MB) This dataset presents the live video streaming sessions of one day (23rd August 2017)

Database: trace.txt (10.7 MB) This dataset presents a simple video streaming delivery in CDN

## Source code

Download our maltab code (9 KB) The original matlab code of all experiments in our paper, it contains 6 matlab (.m) fils

### Files Contents

#### data_divide.m

Analyse the Database: output.txt to get the relationship between each item. The following four programs directly call the variables that are generated in this file.

#### random_algo.m

Calculate the number of delivered segments for different number of users using one edge-server for two implementations: with our proposal and with the randomized approach.

#### CDF.m

Calculate the cumulative Distribution Function (CDF) of the number of users (≤ 300) per edge-server for each video in the CDN.

#### traffic_saving.m

Compare the traffic when the watermarked sequences are distributed with randomized approach and with our proposal, and also calculate the traffic saving ratio with respect to the randomized approach when the videos transmitted through different number of edge-servers.

#### intermediate.m

Calculate the number of watermarked segments that origin server delivers and intermediate servers deliver for two implementations: with our proposal and with the randomized approach.

#### group_improve.m

Analyse another Database: trace.txt. First calculate the best and worst cases of user-server assignment in unicast scenario, then calculate the number of delivered segments for 100 movies when the 12 edge-servers are independent, and when the edge-servers are groups into 6 regions and 4 regions.