Skip to Main content Skip to Navigation

Video conference based on enterprise desktop grid

Abstract : There exist two classical and well-understood approaches to video processing tasks for videoconferencing. The first one is using a centralized Multipoint Control Unit (MCU). In the second approach, the video processing tasks are directly handled in endpoints. Performance is then restricted by device characteristics, especially in the case of mobile devices. In this thesis, we propose a third alternative approach. We propose a system, which distributes real-time video processing tasks on enterprise desktop grid. A dedicated Multi Attribute Decision Making method is designed in order to take into account the variety of attributes impacting Quality of Experience. A number of task distribution and redistribution algorithms are elaborated. We then test the proposed approach by means of simulation in order to study the impact of the main critical parameters. The proposed approach arises a question to which extent a PC can be used as a platform for media server and how CPU load affects the quality of provided video conference. We estimate a perceived quality of video streams in order to investigate CPU load influence. Also we elaborated algorithms, combining Cloud/Fog approach with different types of media servers, the result provides an optimized conferencing solution in the terms of cost for both provider and consumer as well as in terms of end user experience. Combining elaborated algorithms and architecture with experimentation results we conclude that proposed solution can be used as a novel approach to video conferencing problematic.
Complete list of metadata

Cited literature [102 references]  Display  Hide  Download
Contributor : ABES STAR :  Contact
Submitted on : Monday, May 14, 2018 - 1:19:06 PM
Last modification on : Friday, July 31, 2020 - 10:44:09 AM
Long-term archiving on: : Tuesday, September 25, 2018 - 8:57:29 AM


Version validated by the jury (STAR)


  • HAL Id : tel-01791111, version 1


Roman Sorokin. Video conference based on enterprise desktop grid. Distributed, Parallel, and Cluster Computing [cs.DC]. Télécom ParisTech, 2017. English. ⟨NNT : 2017ENST0006⟩. ⟨tel-01791111⟩



Record views


Files downloads