In-network Quality of Experience Monitoring of HTTP Adaptive Streaming Services (CROSBI ID 685093)
Prilog sa skupa u zborniku | izvorni znanstveni rad | međunarodna recenzija
Podaci o odgovornosti
Oršolić, Irena ; Skorin-Kapov, Lea
engleski
In-network Quality of Experience Monitoring of HTTP Adaptive Streaming Services
Due to the increasing adoption of encryption in Over-the-Top services, Internet Service Providers (ISPs) generally lack insight into service performance. Thus, to be able to detect Quality of Experience (QoE) degradations that could potentially be mitigated in the network, ISPs are looking for ways to infer QoE and Key Performance Indicators (KPIs) from statistical properties of network traffic. Research on this topic, including our work, has found machine learning techniques to give promising results for training models that map network-level parameters to QoE/KPIs. While a lot of efforts are currently being put into tackling this challenge, a number of open issues remain. The aim of this research is to further explore the possibilities of this approach and specify a framework that automates the model training process in a service- and platform-independent way, enabling extensive model (re)training and analysis.
QoE monitoring ; encrypted traffic ; YouTube ; machine learning
nije evidentirano
nije evidentirano
nije evidentirano
nije evidentirano
nije evidentirano
nije evidentirano
Podaci o prilogu
1-2.
2018.
objavljeno
Podaci o matičnoj publikaciji
Proceedings of the International Conference on Smart Systems and Technologies 2018 (SST 2018)
Podaci o skupu
International Conference on Smart Systems and Technologies 2018(SST 2018)
poster
10.10.2018-12.10.2018
Osijek, Hrvatska