Big Data Visualization for Crowdsourced Network Measurements

The goal for this project is to build and maintain a platform which allows for the automatic merging, representation, and analysis of multiple crowdsourced network measurement datasets from different sources. The applications include tracing the evolution of network performance during the COVID-19 pandemic.
Master

End-to-end measurements highlight network performance (in terms of "speed" and other metrics) across different network operators, access technologies, regions, time periods, and devices. Crowdsourced applications are widely used to run such measurements.

There are a number of regulatory crowdsourced applications in Europe, where most authorities provide their results as open data. Examples include Austria (RTR-Nettest), Croatia (HAKOMetar Plus), Czech Republic (NetMetr), Slovakia (RU MobilTest), Slovenia (AKOS Test Net), and Norway (NKOM Nettfart). Most of these tools have comparable measurement methodologies.

The exponentially growing datasets from these tools can be merged to create a continuous pool of data, which could be used for the joint analysis of networks from all over Europe, such as tracing the evolution of network performance during the COVID-19 pandemic.

Goal

The idea for this project is to build and maintain a platform which allows for the automatic merging, representation, and analysis of multiple datasets from different sources.

The platform should comprise a scalable backend solution for ingest and storage, a dashboard-like UI for the presentation of statistics, and an easy export functionality for integration with external analysis pipelines (e.g., for ML applications on data time series).

The platform will be used to pre-process and analyze crowdsourced network measurements, e.g., benchmarking network performance across different operators, and building ML models to represent temporal/spatial evolution.

Learning outcome

Crowdsourced measurements, data analysis, machine learning, wireless communications and mobile broadband networks

Qualifications

  • Fixed/mobile broadband networks
  • Software programming (e.g., Python)
  • Database solutions (SQL/noSQL)
  • Visualization solutions (Python Dash, Grafana, Google Data Studio, Tableau, etc.)

Supervisors

  • Pål Halvorsen
  • Michael Riegler
  • Cise Midoglu

References

RTR-Nettest


HAKOMetar Plus


NetMetr

 

RU MobilTest


AKOS Test Net

 

NKOM Nettfart

 

Contact person