OPUS - Publikationenserver der Technischen Hochschule Nürnberg Georg Simon Ohm
Not a member yet
2332 research outputs found
Sort by
Optical Enabling Technologies for Quantum Technologies
No full textAt TH Nürnberg, we are in the process of procuring large-scale research equipment for additive manufacturing of microlenses and deterministic generation of color centers. We present how this equipment shall be used for the manufacturing of optical and photonic interfaces for the miniaturization and integration of quantum systems. Both equipment and cleanroom of TH Nürnberg could be used by MQV and its partners
Eye Scan UX Sound Auditory Icon
No full textDer Augenscan ist ein im Rahmen eines Forschungsmasters mit modularer Klangsynthese erzeugtes Auditory Icon als WAV-Datei.The eye scan is an auditory icon created as part of a research master by using modular sound synthesis (WAV file)
Digital Operating Mode Classification of Real-World Amateur Radio Transmissions
No full textThis study presents an ML approach for classifying digital radio operating modes evaluated on real-world transmissions. We generated 98 different parameterized radio signals from 17 digital operating modes, transmitted each of them on the 70 cm (UHF) amateur radio band, and recorded our transmissions with two different architectures of SDR receivers. Three lightweight ML models were trained exclusively on spectrograms of limited non-transmitted signals with random characters as payloads. This training involved an online data augmentation pipeline to simulate various radio channel impairments. Our best model, EfficientNetB0, achieved an accuracy of 93.80% across the 17 operating modes and 85.47% across all 98 parameterized radio signals, evaluated on our real-world transmissions with
Wikipedia articles as payloads. Furthermore, we analyzed the impact of varying signal durations & the number of FFT bins on classification, assessed the effectiveness of our simulated channel impairments, and tested our models across multiple simulated SNRs
Optimized Self-supervised Training with BEST-RQ for Speech Recognition
No full textSelf-supervised learning has been successfully used for various speech related tasks, including automatic speech recognition. BERT-based Speech pre-Training with Random-projection Quantizer (BEST-RQ) has achieved state-of-the-art results in speech recognition. In this work, we further optimize the BEST-RQ approach using Kullback-Leibler divergence as an additional regularizing loss and multicodebook extension per cluster derived from low-level feature clustering. Preliminary experiments on train-100 split of LibriSpeech result in a relative improvement of 11.2% on test-clean by using multiple codebooks, utilizing a combination of cross-entropy and Kullback-Leibler divergence further reduces the word error rate by 4.5%. The proposed optimizations on full LibriSpeech pre-training and fine-tuning result in relative word error rate improvements of up to 23.8% on test-clean and 30.6% on testother using 6 codebooks. Furthermore, the proposed setup leads to faster
convergence in pre-training and fine-tuning and additionally stabilizes the
pre-training
Machine-Readable by Design: Language Specifications as the Key to Integrating LLMs into Industrial Tools
No full textWe propose a meta-language-based approach enabling Large Language Models (LLMs) to reliably generate structured, machine-readable artifacts referred to as Meta-Language-defined Structures (MLDS) adapted to domain requirements, without adhering strictly to standard formats like JSON or XML. By embedding explicit schema instructions within prompts, we evaluated the method across diverse use cases, including automated Virtual Reality environment generation and automotive security modeling. Our experiments demonstrate that the meta-language approach significantly improves LLM-generated structure compliance, with an 88 % validation rate across 132 test scenarios. Compared to traditional methods using LangChain and Pydantic, our MLDS method reduces setup complexity by approximately 80 %, despite a marginally higher error rate. Furthermore, the MLDS artifacts produced were easily editable, enabling rapid iterative refinement. This flexibility greatly alleviates the “blank page syndrome” by providing structured initial artifacts suitable for immediate use or further human enhancement, making our approach highly practical for rapid prototyping and integration into complex industrial workflows
AI Integration in Business Process Management
No full textThis paper presents issues related to implementing Artificial Intelligence (AI) technology in business process management. The general concept of AI, its key technical characteristics, and functional capabilities that facilitate company activity optimization and automation are discussed. Based on analysis of various companies' experiences already utilizing AI in business management practices, the multifaceted effect of technology implementation emerged, including operational cost reduction, productivity growth, customer satisfaction improvement, and competitive advantage acquisition in the market. The article highlights areas where AI application yields high effectiveness and identifies management processes requiring its integration. The findings demonstrate that AI is a strategic business transformation tool that increases competitive advantage in rapidly changing economic environments
Exploring Hydrogen Technology Adoption in the German Energy Sector — A Dynamic Capabilities Perspective
No full textThe transition to hydrogen is an important element of the sustainable transition of Germany’s industry. Regional German municipal utilities play a key role in this transition, as they are at the core position between local demand and regional supply. They play an important role when it comes to planned infrastructure development on the regional level. As energy infrastructure development is a long-term project, they need to deal with the hydrogen transition at a very early stage, including pilot projects, organizational changes, and business model adaptations. Yet, not much is known about how these entities organize the transition to hydrogen. Using a qualitative research approach, this study takes a dynamic-capability perspective to shed light on the hydrogen transition activities of German municipal utilities. Industry and other energy sector participants can learn
from these insights on how to organize the transition path towards hydrogen. Findings indicate that strong partnerships play a key role in identifying new opportunities and that balancing risk and benefit is a key challenge when it comes to realizing new technological hydrogen projects
Managing the Sustainable Development of Kazakhstan’s Labor Market Through Gender Equality
No full textGender equality is necessary for further economic development of a country and societal welfare in conditions of the modern demands and shifts in the labor market of Kazakhstan. The goal of this research is to reveal the significance of gender issues in the sustainable enhancement of the labor market in Kazakhstan and to suggest the possibilities of applying gender equality in management practices. Regression and correlation analysis were conducted to analyze the relationship of indicators of gender equality with the economic data. A strong positive correlation (r = 0.909, p = 0.000265) was found between the ratio of women’s wages to men’s wages and the proportion of women in economic activity groups, indicating that women’s participation in the labor market is associated with an increase in their wages. The results of the study show that there is specific progress in the labor market of Kazakhstan in relation to gender inequality, but structural barriers remain. To achieve sustainable development, comprehensive measures are needed to ensure wage equality, increase the participation of women in high-paying industries, and create a gender balance in leadership positions. Thus, it is clear that the enhancement of gender equality increases labor productivity, expands personnel stock, and enhances the resilience of the economy. Future research in managing the sustainable development of the labor market in Kazakhstan through gender equality can be aimed at studying the long-term impact of gender initiatives on economic growth and social stability and assessing the effectiveness of specific policies and programs
Centralized SoC Balancing for Batteries with Droop-Controlled DC/DC Converters for Electric Aircraft
No full textIn this article, an approach to balance the State of Charge (SoC) of two batteries connected to the DC bus of a fuel cell (FC) electric aircraft by Droop-controlled converters is described. The proposed algorithm is based on shifting the Droop reference voltages and prevents the simultaneous charging and discharging of the batteries. This approach is not only practical but also highly versatile, as it is compatible with all converters as long as the Droop voltage can be changed remotely, and a current measurement is provided to a central controller. No further programming access to the DC/DCs is necessary. There is no need for nonlinear or different-valued Droop resistances for charging and discharging. The balancing approach is validated via simulation in MATLAB/Simulink 2024a.The results show that the proposed approach achieves SoC balancing without degrading the dynamic performance of the grid. The delays added by the slower communication with the central controller have a minimal impact on performance