Congress & Conference Contributions

In the present paper we introduce the empirical results of measurements with an over-the-air based propagation artifact for verification and validation of sub-THz and THz channel sounders and parameter estimation algorithms. This experiment produces a fixed number of multipath components with traceable propagation properties in the different domains that can be used to test resolution and performance. Because of the inherent characteristics of the measurement hardware, we have introduced an adaptation on a parametric high resolution estimation algorithm to account the imperfections of the channel sounder. The results have shown to account for a relative good performance of the sounder and the tested parametric and non-parametric estimation algorithms.

Ray-tracing methods can be applied to evaluate the wave propagation in anechoic electromagnetic environments, in order to improve measurement concepts, or to address the increasing demands on test concepts for new radio wireless transmission functionalities. The accuracy of the data obtained from such a chamber model depends crucially on the modeling of the microwave absorbers. Often the scattering off the absorbers is simplified by specular reflections, ignoring the actual direction-dependent scattering properties of the absorbers resulting from their geometry, arrangement, and material. As this may cause significant misconceptions, improved ray-tracing models of absorbers based on their scattering character are needed. In this paper, the scattering of two types of 18″ absorbers with pyramidal and wedge geometries is analyzed and expected scattering patterns are described. Based on an existing shooting-and-bouncing-rays software, a simulation approach is introduced and used to relate the analytical findings with results of measurements of the angle-dependent reflectivity. In addition to the primarily specular reflection for wedge absorbers and scattering in many directions off the pyramidal absorbers, the results show that nearfield effects influence this scattering pattern, as eventually confirmed by the simulations.

Channel charting (CC) is an emerging machine learning method for learning a lower-dimensional representation of channel state information (CSI) in multi-antenna systems while simultaneously preserving spatial relations between CSI samples. The driving objective of CC is to learn these representations or channel charts in a fully unsupervised manner, i.e., without the need for having access to explicit geographical information. Based on recent findings in deep manifold learning, this paper addresses the problem of CC via the "not-too-deep" (N2D) approach for deep manifold learning. According to the proposed approach, an embedding of the global channel chart is first learned using a deep neural network (DNN)-based autoencoder (AE), and this embedding is subsequently searched for the underlying manifold using shallow clustering methods. In this way we are able to counter the problem of collapsing extremities - a well known deficiency of channel charting methods, which in previous research efforts could only be mitigated by introducing side-information in form of distance constraints. To further exploit the ever-increasing spatio-temporal CSI resolution in modern multi-antenna systems, we propose to augment the employed AE with convolutional neural network (CNN) input layers. The resulting convolutional autoencoder (CAE) architecture is able to automatically extract sparsely distributed spatio-temporal features from beamspace domain CSI, yielding a reduced computational complexity of the resulting model.

In a wireless sensor network (WSN), sensor nodes are distributed in a predefined environment, collecting environmental data and transferring them to the base station or sink for processing and analysis. In the meantime, much research has been done on heterogeneous wireless sensor networks. Due to the fact that in a heterogeneous wireless sensor network (HWSN) factors such as primary energy, data processing ability, etc. greatly affect the life of the network, the selection of cluster heads (CHs) for coordinating a cluster of sensor nodes has created a wide range for further development of network efficiency capabilities. In this work, an event-driven energy-efficient protocol based on a genetic algorithm is proposed that uses various parameters such as the remaining energy of the node, the minimum distance of a node to the base station, and also the degree of the neighborhood of a node as fitness function values, and evaluates solutions in a heterogeneous wireless sensor network. For evaluating the proposal, network stability, energy consumption, death of the first node, and number of living nodes per round are considered as evaluation criteria.