Adaptive Live Video Streaming Versione del 26 Mag 2019 alle 10:43 di Ldecicco (Discussione | contributi) (diff) ←Older revision | view current revision (diff) | Newer revision→ (diff)

Adaptive Live Video Streaming

Adaptive live streaming represents a key advancement with respect to classic progressive download streaming such as the one employed by YouTube. With adaptive streaming the video source bit-rate is automatically adapted in real-time to match the time-varying bandwidth available to the user who can experience the maximum quality.

Cisco Award

Cisco Award 2013 Funded by "Cisco University Research Program" managed by the Silicon Valley Community Foundation.

• Principal investigator: S. Mascolo
• Title: Architecture for Robust and Efficient Control of Dynamic Adaptive Video Streaming over HTTP.
• March 2013

Best Paper Award

We got the best student paper award at the MMSYS 2016 conference for our work:

• G. Cofano, L. De Cicco, T. Zinner, A. Nguyen-Ngoc, P. Tran-Gia, and S. Mascolo
  Design and Experimental Evaluation of Network-assisted Strategies for HTTP Adaptive Video Streaming
  Proc. ACM Mmsys 2016, Klagenfurt, Austria, May 2016 (Best student paper award) (PDF)

Selected publications

Design of Adaptive Video Streaming control algorithms

• Luca De Cicco, Giuseppe Cilli, Saverio Mascolo
  ERUDITE: a Deep Neural Network for Optimal Tuning of Adaptive Video Streaming Controllers
  10th ACM Multimedia Systems Conference (ACM MMSys '19), Amherst, MA, USA, June 18-21, 2019

• Giuseppe Ribezzo, Giuseppe Samela, Luca De Cicco, Vittorio Palmisano, Saverio Mascolo
  A DASH Video Streaming System for Immersive Contents
  Proc. ACM Multimedia Systems Conference (ACM MMSYS) - Demo, Amsterdam, The Netherlands, June 2018

• G. Cofano, L. De Cicco, T. Zinner, A. Nguyen-Ngoc, P. Tran-Gia, and S. Mascolo
  Design and Experimental Evaluation of Network-assisted Control Strategies for HTTP Adaptive Video Streaming
  ACM Transactions on Multimedia Computing, Communications, and Applications (TOMM) (special issue of Extended Best Paper Candidates of ACM MMSys 2016 and ACM NOSSDAV 2016), March 2017

• L. De Cicco, V. Caldaralo, V. Palmisano, and S. Mascolo
  ELASTIC: a Client-side Controller for Dynamic Adaptive Streaming over HTTP (DASH)
  Proc. of Packet Video Workshop 2013, San Jose, CA, USA, December 2013 (PDF) (Slides:PDF)

Modeling of Adaptive Video Streaming Systems

• Giuseppe Cofano, Luca De Cicco, Saverio Mascolo
  Modeling and Design of Adaptive Video Streaming Control Systems
  IEEE Transactions on Control of Network Systems, vol. 5, no. 1, pp. 548-559, doi: 10.1109/TCNS.2016.2631452, March 2018 (PDF)

• Giuseppe Cofano, Luca De Cicco, Saverio Mascolo
  A Hybrid Model of Adaptive Video Streaming Control Systems
  Proc. of IEEE Conference on Decision and Control (IEEE CDC 2016), Las Vegas, Nevada, USA, Dec 2016 (PDF) (Slides:PDF)

• L. De Cicco, G. Cofano, and S. Mascolo
  A Hybrid Model of the Akamai Adaptive Streaming Control System
  Nonlinear Analysis: Hybrid Systems, Special Issue 'IFAC World Congress 2014', Elsevier, Volume 21, Pages 139–154, DOI 10.1016/j.nahs.2015.12.007, August 2016

• L. De Cicco and S. Mascolo
  An Adaptive Video Streaming Control System: Modeling, Validation, and Performance Evaluation
  IEEE/ACM Transaction on Networking, Volume 22, Issue 2, April 2014 (PDF)

Demo

The following videos are distributed by employing a prototype implementation of the control algorithm described in the paper:

• L. De Cicco, V. Caldaralo, V. Palmisano, and S. Mascolo
  ELASTIC: a Client-side Controller for Dynamic Adaptive Streaming over HTTP (DASH)
  Proc. of Packet Video Workshop 2013, San Jose, CA, USA, December 2013 (PDF) (Slides:PDF)

The video source is adapted by following the stream-switching (or multiple bit-rate) approach: the video source is available at different bit-rates and resolutions and a controller switches from one video version to match the available bandwidth while avoiding playback interruptions and re-buffering events. The figure below shows the architecture of the adaptive video streaming service that we have designed.

Requirements:

• In order to experience the maximum quality (720p) a dual core 2.4GHz processor or better is recommended.