Evaluating Smartphone Accuracy for LTE Power Measurement

Smartphones are today relatively cheap devices that embed a large variety of sensors such as magnetometers or orientation sensors, but also the hardware to connect to most wireless communication technologies such as Wi-Fi, Bluetooth, or cellular networks. For this reason, companies, such as OpenSignal or Tutela use smartphones to make crowd-based measurements of the received power from the cellular infrastructure to help operators manage their infrastructure. However, to the best of our knowledge, the accuracy of such measurements has never been rigorously assessed. The goal of this paper is to assess how accurate are measurements of received power from a 4G (LTE) antenna when performed from a Commercial Off-The-Shelf~(COTS) smartphone in different environments. We first evaluate the granularity and limitations of the Android API that returns the received power. We explore how reliable are the measurements from a mono-polarized antenna in a fully controlled environment. We show that the orientation of the smartphone, the position of the source, and the distance to the source has a significant impact on the accuracy of the measurements. We introduce several calibration techniques based on radiation matrices manipulations and machine learning to calibrate the measurements, that is, to improve the accuracy to less than 5 dBm RMSE compared to a professional equipment. Finally, we explore how reliable are measurements in an outdoor environment, in the context of a multi-polarized antenna.