Humboldt-Universität zu Berlin - Faculty of Mathematics and Natural Sciences - Department of Computer Science

Our Research

The Chair conducts research on reconfigurable computer architectures – from their fundamental design and the tools used to program them, to their practical application in demanding embedded systems. Prof. Hübner has dedicated more than 20 years to this field of research.

Reconfigurable Architectures: Where Flexibility Meets Efficiency

At the heart of our work lies the idea of designing computer architectures that are adaptable rather than fixed. Reconfigurable systems can be adapted to changing requirements at runtime – combining the flexibility of software with the efficiency of specialized hardware. A particular focus is on runtime-reconfigurable processors for embedded systems: architectures capable of changing their structure during operation to optimally adapt to current tasks, environmental conditions, or energy budgets.

Beyond architecture design itself, we develop tools and methods for programming such systems – since it is only through well-designed design flows and compilers that the advantages of reconfigurable hardware become practically usable for application developers.

Applications with Demanding Requirements

Our architectures are in demand wherever low power consumption and high reliability are equally critical. Key application areas include:

• Industrial and process automation – robust, efficient control systems for continuous operation
• Mobility of the future – deployment in automotive and aerospace systems, where safety and energy efficiency are top priorities
• Mobile communications – flexible hardware for the requirements of today's and future mobile standards (5G, 6G)

More broadly, these hardware architectures are used in sensing and actuation.

Edge Computing and AI-Driven Reconfiguration

A growing area of research is edge computing, where computationally intensive AI methods are executed directly on high-performance, reconfigurable processor architectures close to the data source. Beyond this, we are exploring a novel approach: using artificial intelligence methods to control the runtime adaptation of the architectures themselves. This enables systems to learn how to steer their own reconfiguration intelligently and context-dependently. This combination of adaptive hardware and AI-driven self-control is the subject of fundamental, forward-looking research at our chair.

Our teaching

We offer many courses in the area of computer engineering and networking. Topics range from computer architecture and circuit design, wired and wireless networks and communication protocols to security and algorithms and distributed systems.

Our current teaching programme can be found here (German only).

How to find us

Verantwortlich für die Seite: