Login
 Forum
 
 
Thesis topic proposal
 
Markosz Maliosz
Community based forecasting for effective video content caching in mobile networks

THESIS TOPIC PROPOSAL

Institute: Budapest University of Technology and Economics
computer sciences
Doctoral School of Informatics

Thesis supervisor: Markosz Maliosz
Location of studies (in Hungarian): Távközlési és Médiainformatikai Tanszék
Abbreviation of location of studies: TMIT


Description of the research topic:

Kutatási célkitűzések:
Mobile client devices capable of caching are able to prefetch video content before the actual watching time, out of busy hour, to balance the network load. In one hand, there are video contents that are generally popular, they should be available with quick access everywhere and for everyone using central global caches, replicated in local caches as well. On the other hand, many of the videos are watched by a small number of viewers. Between the above two extremes, there is an open research field where the local popularity of the videos can be harmonized with the locality of the mobile users, thus local caches can be optimized. Based on the user preferences a forecasted video content can be prefetched not only in local caches, but also to the client devices directly.
Based on the user profiles and the available content a selective caching method should be developed. The performance of the local selective caches and the prefetching for the client devices can be enhanced by evaluating the behavior of the community of the video sharing sites deciding on what content, when, and to whom to preload. The goals for the selective caching method are effective service provisioning and user satisfaction.
Open problems:
- Analysis and characterization of mobile video content related user profiles
- Determining the harmonization between user locality and popularity groups
- Development of a forecasting algorithm for selective video content caching and prefetching
- Construction of an objective function which decides on whether the content have to be prefetched entirely or only partially
Requirements:
- experience in user data collection methods
- good knowledge in statistics

Required language skills: angol
Number of students who can be accepted: 1

Deadline for application: 2016-01-04


2024. IV. 17.
ODT ülés
Az ODT következő ülésére 2024. június 14-én, pénteken 10.00 órakor kerül sor a Semmelweis Egyetem Szenátusi termében (Bp. Üllői út 26. I. emelet).

 
All rights reserved © 2007, Hungarian Doctoral Council. Doctoral Council registration number at commissioner for data protection: 02003/0001. Program version: 2.2358 ( 2017. X. 31. )