Crowd Matters: Analysis of different crowdsource platforms for mobile performance


Crowdsourcing apps (e.g. mobiperf, speedtest, etc...) are widely used to measure mobile broadband (e.g. 3G, 4G) performance by highlighting performance differences across operators, access technologies, and regions. However, different apps run different tools to measure basic mobile performance which in turn has a significant impact on the conclusions drawn from these measurements. The focus of this thesis is to analyse different crowdsourced datasets in a certain area, identify the differences in the tools and define best practices on how these measurements should be carried out.

Learning outcome

  • Understanding of mobile networks
  • Understanding of crowdsource platforms
  • Understanding of how to measure mobile performance
  • Experience with large data analysis tools and methods


Programming skills (preferably python or R) and analytical skills for data analysis


Özgü Alay