You are here

Mobile network performance issue cause analysis

General description

This use case aims in detecting the existence, location and root cause of video streaming QoE issues on mobile devices. To achieve this, we place measurements probes on the mobile device, the wireless gateway and the content server. Each probe is responsible for collecting performance metrics from different layers such as hardware, TCP, link, OS and application. After the playback of a video, the respective collected metrics are sent to a remote repository. The reasoner then will assess if there were QoE issues with the particular video session, where did the issues occur and what was root cause behind them.

An overview of the use case is shown in the figure below.

Components describing the architecture, and point to pieces of software

The use case architecture includes three components: the probes, the repository and the reasoner.

The probes are installed on three different vantage points (e.g. the mobile device, the router and the server) and are responsible for collecting performance metrics and forwarding them to the repository.

The repository consists of a Mongo Database and a node.js interface and is capable of receiving and storing the measurements from the probes that correspond to unique video sessions.

Finally, the reasoner serves the purpose of simulating a range of different faults that affect video QoE while a video is being streamed. Once the video session is complete, the reasoner produces an output to inform if the selected video suffered QoE issues, where the issues occured and what was their root cause.



How to setup and deploy the use-case  


For the detailed instructions on the deployment, setup, and demo of the use-case we refer interested readers to the corresponding demontration guidelines page (link).




HOWTO setup and run it

In order to deploy and run the use case, it is first required to install and run all the probes at the corresponding vantage points. Once the probes have been installed and initialized, a series of random videos will be streamed from the content server to the mobile device. At the same time the probes will collect and periodically send aggregated measurements to the repository.

The use case is controlled via the reasoner's Fault Simulation Component, where a specific fault is simulated during the streaming of a video. After the completion of the video session, the reasoner will automatically generate a report with the confidence with which QoE issues were detected, their location was detected and the confidence of identifying the underlying root cause.

tests done considering integration, performance, validation tests

The use case will be tested and evaluated for different types of faults and for videos with different characteristics to ensure that the system can perform with equally good accuracy under different conditions and for different inputs.

links to all software used

Mobile Probe

Mobile Network RCA Reasoner