System for Continuous Multi-Dimensional Mobile Network KPI Tracking and Prediction in Drifting Environments

Abstract

The usage of public mobile radio networks is steadily increasing. At the same time, the number of new and future smart city applications that rely on reliable and fast mobile data connections based on public mobile networks is rising. In particular, mission-critical smart city applications require continuous and reliable mobile network connectivity. However, the fulfillment of KPIs is not given at all locations and varies over time. Thus, use-cases like tele-operated driving profit from and, in some cases, even depend on spatiotemporal connectivity data. Indirectly, connectivity data can also be utilized to calibrate and improve network planning approaches for future network technologies, such as classical ray tracing or innovative datadriven channel modeling approaches.Massive data acquisition is needed to cover vast city-wide areas like the city of Dortmund. Therefore, this paper discusses a system that enables a dedicated, continuous and systematic measurement campaign to solve this challenge. These measurements are realized by a fully automated open-source monitoring application deployed in multiple vehicles of the local waste disposal company, enabling continuous and city-wide data collection. The initial results of this measurement campaign indicate that up-to-date data is crucial for reliable data-driven services.