24 April 2025 in IEEE Transactions on Machine Learning in Communications and Networking
Future mobile use cases such as teleoperation rely on highly available mobile networks. Due to the nature of the mobile access channel and the inherent competition, the availability may be restricted in certain initially unknown areas or timespans. We automated mobile network data acquisition using a smartphone application and dedicated hardware to address this challenge, providing detailed connectivity insights. DoNext, a massive dataset of 4G and 5G mobile network data and active measurements, was collected over two years in Dortmund, Germany. To the best of our knowledge, it is the most extensive openly available mobile dataset. Machine learning methods were applied to the data to demonstrate its utility in key performance indicator prediction. Radio environmental maps facilitating key performance indicator predictions and application planning across different locations are generated through spatial aggregation for in-advance predictions. We also showcase signal strength modeling with transfer learning for arbitrary locations in individual mobile network cells, covering private and restricted areas. By openly providing the dataset, we aim to enable other researchers to develop and evaluate their machine-learning methods without conducting extensive measurement campaigns.
10 June 2026 in IEEE Access
Future Cooperative Intelligent Transport Systems (C-ITSs) will impose stringent requirements on mobile networks in terms of latency, data rate, and reliability. These requirements must be met in rapidly changing wireless channel conditions. Hence, their fulfillment requires rigorous and reproducible testing environments that enable a seamless transition from simulation to real-world deployment. However, existing testbeds and simulation environments either lack physical realism or fail to scale cost-effectively, hindering comprehensive C-ITS evaluation. To overcome these limitations, we implement the concept of a Digital Triplet, comprising three complementary layers: a full-scale real-world test field, a co-simulation environment for integrating additional simulated entities, and a scaled-down replica of the test field. This architecture enables reliable, reproducible, and scalable testing of future 6G-enabled C-ITS applications across different levels of abstraction. We demonstrate the applicability of the Digital Triplet through multiple case studies. For remote driving, we deploy and optimize a low-latency video streaming system and simulate a mmWave communication channel incorporating Reconfigurable Intelligent Surfaces (RISs), and achieve up to a 74% reduced median Glass-To-Glass (G2G) latency compared to related work. Moreover, emulating real-world vehicle localization in the scaled test field enables the implementation of cross-domain Nonlinear Model Predictive Control (NMPC)-based vehicular platooning. Finally, we evaluate the NMPC platooning controller in both decentralized (on-vehicle) and centralized (edge cloud) configurations. The results highlight the strong potential of the Digital Triplet to accelerate the development and validation of C-ITS applications within future 6G networks.