Filters

Settings

Now showing1 - 10 of 21
  • Some of the metrics are blocked by your 
    consent settings
    Item type:Project, Access status: without files ,
    Fahrplanoptimierung durch eine KI-basierte Lösung zur besseren Prognosefähigkeit im Flow-Management des Personentransfers
    FPOplus ist ein Verbundprojekt zur Entwicklung einer KI-basierten Lösung zur besseren Prognosefähigkeit im Flow-Management des Personentransfers in Echtzeit durch Integration und Transfer innovativer Funktechnologie- und Vorhersagemodelle. Es ist gefördert durch die Ausschreibung Innovative Netztechnologien (InnoNT) des Bundesministeriums für Digitales und Verkehr (BMDV) und wird gemeinsam durchgeführt mit den Verbundpartnern HafenCity Universität Hamburg (HCU), Universität zu Lübeck (UzL), Rheinland-Pfälzische Technische Universität Kaiserslautern Landau (RPTU) und in Zusammenarbeit mit Breuer Nachrichtentechnik GmbH und Deutsche Bahn AG.
      109
  • Some of the metrics are blocked by your 
    consent settings
    Item type:Project, Access status: without files ,
      55
  • Some of the metrics are blocked by your 
    consent settings
    Item type:Project, Access status: with files ,
    VEREDUS
    For many years now, the operation of Unmanned Aerial Vehicles (UAVs) is deemed to revolutionize current transportation, monitoring and surveillance application and enable many other use cases. For this, the safe and reliable operation of UAVs is a necessity. The goal of the VEREDUS project is to connect UAVs among each other and to ground stations so that a precise picture of the UAV positions in the airspace is available at all times. For this purpose, two communication networks are to be established: On the one hand, direct communication allows the distribution of position reports as well as their verification by the means of triangulation. On the other hand, a Low Earth Orbit (LEO) satellite system is used for a redundant control connection to increase the reliability of the system. To realize the VEREDUS system, first, the requirements as well as the operational scenarios are defined. Based on that, the structure of the network is developed and fitting technologies are selected. Special purpose protocols for the distribution of position reports and for multi-hop communication are designed. The performance of the VEREDUS system is analytically modelled using stochastic mobility, signal flow graphs as well as queueing theory. A system-level simulator is implemented to evaluate the system in concrete scenarios. The project is concluded by building the actual hardware, mounting it on UAVs and demonstrating the system in operation.
      136  18
  • Some of the metrics are blocked by your 
    consent settings
    Item type:Project, Access status: without files ,
      136
  • Some of the metrics are blocked by your 
    consent settings
    Item type:Project, Access status: without files ,
      2219
  • Some of the metrics are blocked by your 
    consent settings
    Item type:Project, Access status: without files ,
      123
  • Some of the metrics are blocked by your 
    consent settings
    Item type:Project, Access status: without files ,
      153
  • Some of the metrics are blocked by your 
    consent settings
    Item type:Project, Access status: without files ,
      123
  • Some of the metrics are blocked by your 
    consent settings
    Item type:Project, Access status: without files ,
    I³-Project - Machine Learning in Aeronautical Communications
    The research area of Machine Learning provides methods and data-driven statistical models that "learn" through experimentation or observation. This is advantageous for complex problems where it becomes very difficult if not impossible to design exact models. As a consequence, deep learning achieves breakthroughs mainly in areas where models do not exist at all or human-made models capture only small parts of the reality. The highly dynamic problem space of aeronautical communications in a shared radio spectrum is such a field. While Machine Learning-based methods promise good performance, the common pitfall is that such algorithms provide a well-performing blackbox. In aeronautical communications, however, strict performance guarantees have to be made and kept. In this project, the Institute of Communications and the Institute of Communication Networks will join forces to turn Machine Learning-based methods into a scientifically rigorous approach. Machine Learning-based methods are to be applied, evaluated and explained to problems in the field of communications engineering. These shall be shown to work within strict performance bounds required for safety-critical applications. For this, expert domain knowledge is to be combined with data-driven learning algorithms. The understanding of internal processes is focused upon, and the information bottleneck method shall be applied to gain understanding of the intricacies of deep learning methods. At the same time, novel efficient signal processing units shall be designed, leveraging the information bottleneck method, to develop deep learning solutions that are practically implementable in hardware-constrained communication devices. The ongoing cooperation with scientific and industrial partners as well as the preparation of a project application at the DFG are key objectives of this endeavor.
      389
  • Some of the metrics are blocked by your 
    consent settings
    Item type:Project, Access status: without files ,
    Optimal Utilization of Renewable Energies in Low Voltage (LV) Power Distribution Systems
    The increasing amount of distributed renewable energy resources (DER) installed in low voltage power distribution grids (LVDG) imposes new challenges on the grid operation. Load flows will not be predictable anymore and a monitored and managed grid operation will become necessary. Many DER as well as new types of electric loads increase the risk of critical load flows. As the extension of grid capacity is costly and inefficient, new control algorithms are needed to manage the distributed power production and consumption within LVDG, aiming at maximum usage of DER while keeping grid parameters within their limits.Therefore, a utility-based optimization is proposed, a concept that is common practice in com-munication networks, and applied to the problem of determining optimal power generation of DER and consumption of controllable loads within an LVDG. Optimality is understood as maximum utility experienced by producers and consumers. While the optimization algorithm itself shall be ignorant of constraints on grid currents and voltages, it is closely coupled to a corrective power flow module that identifies critical situations and uses sensitivity analysis to propose corrections on single units’ power generation or consumption.At first, a solution of the static problem is implemented at a central place, meaning that power level bounds and producer/consumer utility functions remain constant during the optimization process. This shall demonstrate the applicability of the concept.The final goal is to develop a distributed optimization algorithm under dynamically changing pa-rameters. It shall run at a high update rate, applying incremental power level changes towards the optimum in each time step. A state estimation in distribution grids is employed for identification of the actual grid state. The high update frequency presumably allows for linearized system modelling. The resulting influences on convergence and stability of the optimization algorithm are main questions to be addressed. The distributed nature, which is also well known from communication networks, is expected to increase robustness and ensure scalability while limiting computing demands on single devices. These properties are crucial for reliable and economically efficient operation of a dynamically changing LVDG. At the same time, the distributed approach requires extensive exchange of information between the participants, putting a high burden on the communication network. The performance of the communication network therefore has a crucial impact on the performance of the LVDG optimization algorithm. For the evaluation of the applicability of the distributed algorithm and for the investigation of the overall performance of the proposed utility-based approach, a model of the communication networks is established. This model mainly focuses on the delay distribution functions using mathematical models such as Signal Flow Graphs, Stochastic Network Calculus and Queueing Theory.
      1236