It is foundamental in the Internet era to understand what affects the Quality of Experience and optimize users' satisfaction.
Since Internet Service quality of experience is intrisicaly related to the network state it is possible to model a function that takes as input network configurations and returns the QoE.
Our first proposal use such kind of functions to give to the user a simple way to understand the expected performances of some popular Internet Services based on it's connectivity.
A classical Skype call can easily turn into this plot.
This is annoying since end users do not have intuitive ways to troubleshoot their experience using Internet Services.
Moreover there are no tools that allow to predict the final QoE of an application without running the application itself.
Performing active measurements to sense the network state.
Using a general measurement plane to track the final user experience.
With only network measurements it is possible to retrieve the QoE of multiple Internet Services. To do so we use several models independent of any application-level details.
The network conditions will be simulated using DummyNet, a live network simulation too.
Our ACQUA prototype will show how the prediction is calculated
in real time using the enforced network conditions.
The expected QoE output is computed from the model in use on
our client; to validate the output prediction just make a call with
Skype!
Thierry Spetebroot, Salim Afra, Nicolas Aguilera, Damien Saucez, Chadi Barakat, "From network-level measurements to expected Quality of Experience: the Skype use case“, in proceedings of the IEEE International Workshop on Measurement and Networking (M&N), Coimbra, Portugal, October 2015.