The advent of 5G and beyond systems is expected to shape the automotive vertical, as safety-critical vehicular applications rely on the network to meet their stringent Quality of Service (QoS) requirements. Predictive QoS (pQoS) has been proposed as a mechanism that allows automotive applications to proactively adapt in view of forthcoming QoS changes. Although pQoS is typically facilitated via classical (centralized) Machine Learning (ML) methods, the demand for data privacy has led to the emergence of distributed ML schemes. Efficient training of ML models however requires large volumes of (kinematic-state and connectivity) QoS data, so as to capture the involved spatio-temporal effects.To that end we hereby present and publicly share NordicDat, a QoS dataset collected during a two-week measurement campaign, driving across three European countries. NordicDat contains over 90K samples of physical layer, network and mobility-related features. Contrary to prior works, it includes multiple instances of cross-boarder roaming, diverse vehicle speed profiles and radio access technologies (generations). Further, we provide a thorough NordicDat data analysis, highlighting the dependencies between the NordicDat’s features and the resulting QoS values (throughput, delay). To showcase its broad usability, we train pQoS ML models over NordicDat in classical and distributed fashion. Our results demonstrate for the first time the viability of distributed pQoS with real-word data, which achieves similar (within a margin of 10%) accuracy to that of classical ML, cropping privacy-preserving benefits.

The successful application of augmented reality as a guidance tool for procedural tasks like maintenance or repair requires an easily usable way of modeling support processes. Even though some suggestions have already been made to address this problem, they still have shortcomings and don't provide all the required features. Thus in a first step the requirements a possible solution has to meet are collected and presented. Based on these, the augmented reality process modeling language (ARPML) is developed, which consists of the four building blocks (i) templates, (ii) sensors, (iii) work steps and (iv) tasks. In contrast to existing approaches it facilitates the creation of multiple views on a single process. This makes it possible to specifically select instructions and information needed in targeted work contexts. It also allows to combine multiple variants of one process into one model with only a minimum of redundancy. The application of ARPML is shown with a practical example.

In this paper, a novel 42 slot, 20 pole ironless winding for use in an axial-flux machine is presented. The proposed design improves the utilization of the available air gap area and is based on the implementation of the winding as a printed circuit board (PCB). Besides, it is compared with a conventional 21 slot, 20 pole ironless axial-flux winding regarding the criteria: torque generation and induced voltage at constant speed. The comparison is based on analytical calculated results, which were obtained by using the winding function theory and an exponential function approach for the magnetic flux density produced by the permanent magnets. The analytical description of the MMF is derived from the geometry whereas the torque and the induced EMF are calculated by using complex Fourier series. Furthermore, a special case of Parseval's theorem is used to get the torque and voltage from the Fourier coefficients in a very simple way. The analytical calculated results are compared to those obtained by FEM.