Generierung von Elektrizitätslastkurven für Private Haushalte - ein Vergleich von Bottom-Up, Top-Down und GAN Ansätzen
Akkurate häusliche Lastkurven sind essenziell für diverse Anwendungen. Aufgrund von Datenschutzbestimmungen sind hochauflösende Lastkurven jedoch sehr selten öffentlich verfügbar. ...
LogInsight: A Scalable, Multiformat Log Analysis Framework for System Anomalies Detection
Modern IT systems and networks are complex and greatly diverse. To maintain their operation and resilience against malicious attacks, security teams invoke to the log analysis. A log is a notebook that summari...
Systematische Analyse von Planungsalgorithmen in 5G-TSN-Netzwerken
Die Integration von 5G und Time-Sensitive Networking (TSN) gilt als Schlüsseltechnologie für mobile und flexible Szenarien der Industrie 4.0, doch die Umsetzung scheitert häufig am Konflikt zwischen stochastischer ...
Fairness by Design in Shared-Energy Allocation Problems: Modellierung, Simulation und Erweiterung
Die Integration erneuerbarer Energien führt zu höherer Volatilität in Stromnetzen und erfordert neue Kooperationsmechanismen zwischen Verbrauchern und Prosumern. Microgrids stellen hierfür eine vielver...
A framework for preparing and assessing cybersecurity datasets
Cyber-physical security research relies on datasets that are trustworthy, well-documented, and temporally coherent. In practice, many public datasets often show weak documentation, inconsistent or missing labels, unstable feature defini...
Enhancing Cybersecurity: LLM-based Intrusion Detection System.
The threat of cyberattacks is challenging modern digital infrastructure. While traditional intrusion detection systems are already effective, they have certain limitations, for example, in detecting new attack patterns. With ...
Dynamic Anomaly Detection for Evolving Cyber Threats Using Adaptive LSTM Networks
Cybersecurity systems must adapt to constantly evolving threats, where traditional static models, such as LSTM, struggle with concept drift and unseen attack vectors. This thesis proposes an adaptive anomaly...
A Transformer-Based Approach to Intrusion Detection in IoT Networks
The Internet of Things (IoT) is transforming industries like healthcare and industrial automation, while becoming a key part of everyday life through smart home devices. However, the complex and interconnected nature of ...
Towards an Evaluation Framework for Co-Simulation Middlewares
The integration of complex, multi-domain models through co-simulation frameworks is essential for the analysis of modern cyber-physical systems. However, the landscape of available tools presents a critical challenge: prominent framework...
