QoE Estimation in Encrypted Traffic Bogdan Iacob Technical - - PowerPoint PPT Presentation

▶
qoe estimation in encrypted traffic
SMART_READER_LITE
LIVE PREVIEW

QoE Estimation in Encrypted Traffic Bogdan Iacob Technical - - PowerPoint PPT Presentation

Jun 24, 2023 455 likes •660 views

Chair of Network Architectures and Services Department of Informatics Technical University of Munich Streaming Video Detection and QoE Estimation in Encrypted Traffic Bogdan Iacob Technical University of Munich Department of Informatics

slide-1
SLIDE 1

Chair of Network Architectures and Services Department of Informatics Technical University of Munich

Bogdan Iacob Technical University of Munich Department of Informatics Chair of Network Architectures and Services Garching bei München, 13. Oktober 2017

Streaming Video Detection and QoE Estimation in Encrypted Traffic

slide-2
SLIDE 2

1. Introduction 2. Streaming Video Detection 3. QoE Estimation 4. Conclusions 5. Bibliography

2 Bogdan Iacob | Streaming Video Detection and QoE Estimation in Encrypted Traffic | 13.10.2017

Outline

slide-3
SLIDE 3

What is Quality of Experience (QoE)?

  • A metric to define the overall level of customer satisfaction
  • Emerged from Quality of Service (QoS)
  • A key matter in the field of telecommunications
  • Expresses user satisfaction both objectively and subjectively
  • Based on 3 classes of influence factors:
  • Human Influence Factors
  • System Influence Factors
  • Context Influence Factors

3 Bogdan Iacob | Streaming Video Detection and QoE Estimation in Encrypted Traffic | 13.10.2017

Introduction

slide-4
SLIDE 4

Why is the detection of encrypted videos important for QoE?

  • Videos have a considerable impact on the user experience
  • Cisco estimates that by 2021, video streaming will reach 82% of all Internet traffic*
  • Over The Top (OTT) providers, like YouTube and Netflix, already switched to

encrypted traffic

  • Current QoE measurements still rely on deep packet inspection (DPI)

*Cisco Visual Networking Index: Forecast and Methodology, 2016-2021. June 6, 2017.

4 Bogdan Iacob | Streaming Video Detection and QoE Estimation in Encrypted Traffic | 13.10.2017

Introduction

slide-5
SLIDE 5

DNS Lookup

  • Idea: identify video packets by knowing in advance the specific server IP addresses

(research focuses on YouTube servers)

  • The video server IP list is built and constantly updated
  • A video server’s IP is identified by searching for a specific string “r*.googlevideo.com” in

DNS responses or SSL/TLS handshake packets

5 Bogdan Iacob | Streaming Video Detection and QoE Estimation in Encrypted Traffic | 13.10.2017

Streaming Video Detection

slide-6
SLIDE 6

Fingerprinting YouTube Traffic

  • Idea: leverage the 2 streaming modes currently used by YouTube:

1. Apple HTTP Live Streaming (HLS) 2. MPEG Dynamic Adaptive Streaming over HTTP (DASH)

  • HLS has following characteristics:

 Each streaming session starts with the transfer of an index file (“manifest”), which is located on a different server than the associated video files  The ClientHello messages of the SSL/TLS handshakes contain the string “manifest.googlevideo.com”  Right after the handshake, a large video block is sent.

6 Bogdan Iacob | Streaming Video Detection and QoE Estimation in Encrypted Traffic | 13.10.2017

Streaming Video Detection

slide-7
SLIDE 7

Fingerprinting YouTube Traffic

  • Idea: leverage the 2 streaming modes currently used by YouTube:

1. Apple HTTP Live Streaming (HLS) 2. MPEG Dynamic Adaptive Streaming over HTTP (DASH)

  • DASH has following characteristics:

 Each streaming session starts with the transfer of an index file (“initsegment”)  Both the index file and its associated video files are located on the same server  Right after the handshake, a small 1.5KB segment is sent by the video server

7 Bogdan Iacob | Streaming Video Detection and QoE Estimation in Encrypted Traffic | 13.10.2017

Streaming Video Detection

slide-8
SLIDE 8

Fingerprinting Netflix Traffic

  • Netflix encodes the videos as variable bitrate (VBR) and streams them using DASH
  • The combination of VBR and DASH can produce unique fingerprints for each video
  • This pattern can be particularly observable for Netflix videos as they have a higher

amplitude of the bitrate variation, compared to other streaming services

8 Bogdan Iacob | Streaming Video Detection and QoE Estimation in Encrypted Traffic | 13.10.2017

Streaming Video Detection

Figure 1: Netflix video overhead due to HTTP headers and TLS

slide-9
SLIDE 9

Fingerprinting Netflix Traffic

  • Used Tools:

 Adudump: to infer the size of application data units, using the TCP sequence and acknowledgement numbers.  OpenWPM: framework based on Firefox and Selenium, which visits sequentially all URLs offered as input.

9 Bogdan Iacob | Streaming Video Detection and QoE Estimation in Encrypted Traffic | 13.10.2017

Streaming Video Detection

Figure 2: adudump trace. Figure 3: Cumulative probability of identifying a video before a specified amount of time has elapsed.

slide-10
SLIDE 10

Key Performance Indicators – Initial Delay

  • Time span between a user’s video request and the actual playback begin
  • Composed of 2 delays:
  • Network delay (time required to send request to server and receive first segments)
  • Initial buffering delay (time required to fill the initial buffer)
  • Impact on QoE: low

10 Bogdan Iacob | Streaming Video Detection and QoE Estimation in Encrypted Traffic | 13.10.2017

QoE Estimation

slide-11
SLIDE 11

Key Performance Indicators – Stalling

  • Occurs when the content consumption rate exceeds the average download rate
  • Severity of stalls is influenced by 2 factors:
  • Duration
  • Frequency
  • Impact on QoE: high

11 Bogdan Iacob | Streaming Video Detection and QoE Estimation in Encrypted Traffic | 13.10.2017

QoE Estimation

Figure 4: Stalling vs. initial delay for YouTube QoE

slide-12
SLIDE 12

Key Performance Indicators – Average Representation Quality

  • Average quality of all streamed video segments
  • Overall media throughput, measured in bits per second
  • Relevant for video streams which use HTTP Adaptive Streaming
  • Impact on QoE: high

12 Bogdan Iacob | Streaming Video Detection and QoE Estimation in Encrypted Traffic | 13.10.2017

QoE Estimation

slide-13
SLIDE 13

Key Performance Indicators – Representation Quality Variation

  • Changes in the representation quality throughout a video session
  • Based on 2 components:

 Frequency of quality changes  Amplitude of quality changes

  • Impact on QoE: high

13 Bogdan Iacob | Streaming Video Detection and QoE Estimation in Encrypted Traffic | 13.10.2017

QoE Estimation

Figure 5: Frequency and time on highest layer changed simultaneously.

slide-14
SLIDE 14

Machine Learning

  • Idea: use machine learning to quantify the correlation between QoS and QoE
  • The process consists of 3 steps:

1. Data collection – application-level data and network traffic are captured 2. Data processing – traffic features are derived for each video to form datasets 3. Model building – datasets are used to train the model

  • A quality level (high, medium, low) is assigned to each video

14 Bogdan Iacob | Streaming Video Detection and QoE Estimation in Encrypted Traffic | 13.10.2017

QoE Estimation

slide-15
SLIDE 15

Machine Learning

  • Idea: use machine learning to quantify the correlation between QoS and QoE

15 Bogdan Iacob | Streaming Video Detection and QoE Estimation in Encrypted Traffic | 13.10.2017

QoE Estimation

Figure 6: Approach for QoE classication based on network features.

slide-16
SLIDE 16
  • QoE is an important metric used to define the level of customer satisfaction by mobile
  • perators and internet service providers
  • Most of the Internet traffic is video traffic
  • The quality of video streaming has a high impact on the perceived QoE
  • The trend for OTT providers goes towards the use of encrypted traffic
  • Encrypted video traffic can be detected based on the streaming mechanism (DASH/HLS)
  • QoE can be estimated using Machine Learning by leveraging the QoS – QoE correlation

16 Bogdan Iacob | Streaming Video Detection and QoE Estimation in Encrypted Traffic | 13.10.2017

Conclusions

slide-17
SLIDE 17
  • G. Dimopoulos, I. Leontiadis, P. Barlet-Ros, and K. Papagiannaki. Measuring video qoe

from encrypted traffic. In Proceedings of the 2016 Internet Measurement Conference, pages 513-526, 2016.

  • I. Orsolic, D. Pevec, M. Suznjevic, and L. Skorin-Kapov. A machine learning approach to

classifying youtube qoe based on encrypted network traffic. Multimedia Tools and Applications, pages 1-35, 2017.

  • W. Pan, G. Cheng, H. Wu, and Y. Tang. Towards qoe assessment of encrypted youtube

adaptive video streaming in mobile networks. In 2016 IEEE/ACM 24th International Symposium on Quality of Service (IWQoS), pages 1-6, 2016.

  • A. Reed and M. Kranch. Identifying https-protected Netflix videos in real-time. In

Proceedings of the Seventh ACM on Conference on Data and Application Security and Privacy, pages 361-368, 2017.

  • T. Hossfeld, M. Seufert, C. Sieber, and T. Zinner. Assessing eect sizes of inuence factors

towards a qoe model for http adaptive streaming. In 2014 Sixth International Workshop on Quality of Multimedia Experience (QoMEX), pages 111-116, 2014.

17 Bogdan Iacob | Streaming Video Detection and QoE Estimation in Encrypted Traffic | 13.10.2017

Bibliography

slide-18
SLIDE 18

18 Bogdan Iacob | Streaming Video Detection and QoE Estimation in Encrypted Traffic | 13.10.2017

Questions?