Next Generation Networks (NGN)
The ability to communicate is the foundational pillar of modern digital services. Today, every application leverages diverse communication frameworks—whether to optimize resource efficiency in the cloud, connect global users, or retrieve data from remote servers. To achieve this, applications tap into an expansive ecosystem of technologies, ranging from high-speed access networks like 5G and Wi-Fi to specialized transport protocols for seamless data streaming. Driven by economic pressures and the technical maturity of cloud-native architectures, modern services are now designed for extreme scalability. This evolution ensures that an application can be dynamically deployed to serve anywhere from ten to a million users while maintaining consistent, high-quality performance.
The Next Generation Networks working group specializes in the evaluation, modeling, and performance analysis of network technologies across all involved layers, from access, over WAN, to application specific deployments within diverse communication ecosystems. Our methodology follows a rigorous scientific pipeline: monitoring, modeling, analysis, and optimization. By monitoring networks across cloud infrastructures mobile, Internet, and data center environments, we extract targeted insights to identify system interdependencies and optimization potential. Leveraging the long-standing core competencies of the Chair of Communication Networks, the NGN working group transforms these empirical observations into abstract models and simulations, enabling high-fidelity performance forecasting and optimization.
Monitoring is an essential first step towards understanding and improving any system. It serves to collect and gather the relevant information to develop system models. In the course of increasingly complex and demanding infrastructures across a variety of communication technologies, intelligent monitoring is crucial, which reflects in a key scientific research focus of the group. The research question here is how to implement efficient monitoring for specific systems and how it can be carried out under certain constraints and conditions. These insights are used to both develop novel methodologies for active measurement studies and improve continuous monitoring tasks in real-world, production environments as well as under lab conditions.
Building on the data collection phase, model development serves as the analytical bridge that transforms raw monitoring data into actionable information. By abstracting complex communication infrastructures, or any other system, models provide a formal representation of system behavior, enabling, among others, the investigation of performance, scalability, and resilience aspects. The primary research focus here lies in developing mathematical and computational frameworks that can accurately capture the dynamics of heterogeneous networks while remaining computationally feasible.
The core research question addresses how to balance model fidelity with abstraction levels to ensure predictive accuracy under diverse operating conditions. These modeling efforts are instrumental in simulating "what-if" scenarios, optimizing resource allocation, and predicting future system states, ultimately enabling the design of more resilient, high-performance, and sustainable architectures in both experimental and production settings.
Building on the insights gained from monitoring and the frameworks established through modeling, analysis and optimization represent the final, critical phase of the technical workflow. This stage focuses on diagnosing performance bottlenecks and systematically enhancing system efficiency through data-driven decision-making. The group’s research in this area explores how to leverage a wide range of methodologies to derive meaningful patterns from complex datasets and translate them into optimal configuration strategies.
The central research challenge is to develop robust algorithms that can navigate multi-objective trade-offs, such as balancing throughput against energy consumption, under dynamic and unpredictable workloads. These methodologies are applied to refine existing infrastructures and to enable the tuning of parameters in real-world production networks, ensuring that systems not only meet current demands but are also optimized for future scalability and reliability.
Core Research Areas
We investigate the transition toward 6G, focusing on network convergence and reliability.
-
Time-Sensitive Networking (TSN): Integrating 5G with TSN to support industrial real-time applications "over the air."
-
Network Softwarization: Evaluating the performance of SDN and NFV architectures, specifically hardware-independent data planes and virtualized network functions (VNFs).
-
Signaling Traffic: Large-scale analysis of mobile signaling in global roaming architectures to prevent outages and improve efficiency.
A major pillar of our work is the optimization of Low-Power Wide-Area Networks for smart environments.
-
Convergent IoT Infrastructures: Developing systems that allow IoT devices to dynamically switch between technologies (5G, LoRa, WiFi) to optimize for power or performance.
-
Performance Modeling: Utilizing simulative studies and queuing theory to analyze channel access methods and network infrastructures
-
Energy Efficiency: Quantifying and optimizing the energy footprint of mobile technologies (e.g. NB-IoT, LTE-M, 5G) for a wide range of applications and use cases.
We leverage Machine Learning (ML) and Data Science to create "self-aware" networks.
-
Large-Scale Monitoring: We are constantly working with mobile operators, network providers, and industry partners to gather and evaluate real-world monitoring data from actual production deployments
-
Model Development: The development of a wide range of models (e.g. queueing theory models, simulations, digital twins) is a key step towards understanding and improving modern communication systems across all domains, from enterprise networks, over cloud deployments, to the global mobile ecosystem and non-terrestrial communication
-
Anomaly Detection: Optimizing any system involves understanding what its natural behavior is. To this end, the detection of anomalies, outages, failures, or misconfigurations is critical. We develop mechanisms to detect and classify such outages based on real-world monitoring data as well as synthetic data generated in the lab.
Team
Dr. Stefan Geißler
Head of Research Group
Doctoral Researcher
Doctoral Researcher
Doctoral Researcher
Doctoral Researcher
Doctoral Researcher
Doctoral Researcher
Doctoral Researcher
Current Projects
Serverless Scientific Computing for Earth Observation and Sustainability Research (SOS)
(January 2025 – December 2028)
This project presents an interdisciplinary approach to the development of a large-scale data- and compute-aware workflow-management-system for applications in Earth Observation.
Sustainable Technologies for Advanced Resilient and Energy-Efficient Networks (SUSTAINET)
(January 2025 – December 2027)
This project proposes a holistic approach, integrating research in frictionless network performance, resilience, security, and sustainability to propel Europe towards a sustainable, technologically sovereign future.
Optimized Resource Integration and Global Architecture for Mobile Infrastructure for 6G (ORIGAMI)
(January 2024 - January 2027)
The ORIGAMI project brings together leading players from industry and academia in the mobile telco ecosystem in Europe, and is funded by the European Commission as part of the 6G Smart Networks and Services Joint Undertaking (SNS JU). During the 2023-2025 time frame, the ORIGAMI consortium works towards advancing the architectural models of next-generation mobile networks and removing important barriers towards successful 6G technologies.
NaSA-OMI - Das Netz-als-Sensor: Analyse des Kernnetzes konvergenter Mobilfunksysteme zur Entwicklung neuer Geschäftsmodelle und optimierter Network Intelligence
(November 2025 – December 2028)
The NaSA-OMI project uses Network Intelligence (NI) to detect and analyze issues in global mobile networks, such as faults, malware, and inefficiencies, and to develop intelligent mechanisms for monitoring and improving network performance, enabling new value-added services. Driven by IoT growth and emerging satellite connectivity (GEO/LEO), the project explores integrating advanced analytics and new connectivity technologies to enhance global mobile communication systems.
Modellierung und Monitoring von Virtual Private Cloud Netzen für automatisierte Anomalieerkennung für Unternehmensanwendungen in heterogenen Netzen (VIPNANO)
(October 2023 - September 2026)
In many companies, the demand for flexible and reliable Virtual Private Clouds (VPCs) is increasing, often combining multiple major cloud providers, private clouds, and proprietary networks within a heterogeneous infrastructure. This complexity presents new challenges for network monitoring to ensure reliable operation. This project aims to develop methods for automatically modeling the normal state of traffic patterns and building an automated anomaly detection system on top of it. This will enable the early identification of critical situations in network and application operations, allowing appropriate countermeasures to prevent issues and outages. The project partners bring together the necessary expertise: Isarnet specializes in analyzing network monitoring data, JMU contributes expertise in modeling and simulation, and the enterprise customers, as associated partners, provide large-scale network infrastructure and years of experience in operating applications within VPCs.
