[Back]

C. Midoglu, P. Svoboda:
"Opportunities and Challenges of Using Crowdsourced Measurements for Mobile Network Benchmarking A Case Study on RTR Open Data";
Talk: SAI Computing Conference (SAI), 2016, London; 07-13-2016 - 07-15-2016; in: "Proceedings of the 2016 SAI Computing Conference (SAI)", IEEE (ed.); IEEE, CFP16SAA-USB (2016), ISBN: 978-1-4673-8460-5; 996 - 1005.

Crowdsourcing is a novel paradigm which has applications in a plethora of disciplines including software development, public administration, and communication technologies. In the field of telecommunications, crowdsourcing makes a viable addition to state of the art benchmarking methods for mobile networks, when combined with mobile sensing approaches to employ smartphones as end nodes. In this paper, we review the opportunities and challenges of using crowdsourced measurements from smartphones for benchmarking mobile networks, and demonstrate some of these generic aspects using an open data set containing over two million entries. We show that there is a big potential to distributing performance measurements towards the peripherals of the network, but in order to achieve accurate benchmarking, the collection of data has to be complemented with appropriate signal processing. Our study also emphasizes the importance of open data and open source tooling in achieving fairness and repeatability.

Crowdsourcing is a novel paradigm which has applications in a plethora of disciplines including software development, public administration, and communication technologies. In the field of telecommunications, crowdsourcing makes a viable addition to state of the art benchmarking methods for mobile networks, when combined with mobile sensing approaches to employ smartphones as end nodes. In this paper, we review the opportunities and challenges of using crowdsourced measurements from smartphones for benchmarking mobile networks, and demonstrate some of these generic aspects using an open data set containing over two million entries. We show that there is a big potential to distributing performance measurements towards the peripherals of the network, but in order to achieve accurate benchmarking, the collection of data has to be complemented with appropriate signal processing. Our study also emphasizes the importance of open data and open source tooling in achieving fairness and repeatability.

Smartphone, Benchmarking, Crowdsourcing, Mobile networks, Performance evaluation, QoS, RTR Open Data

http://dx.doi.org/10.1109/SAI.2016.7556101

http://publik.tuwien.ac.at/files/PubDat_251333.pdf