Communication Systems

Dr. Marco Breiling (51) wants to assist in shaping the future of Germany. The electrical engineer from Erlangen helps get successful high-tech products made in Germany. Inspired by the energy efficiency of a bee’s brain, his team develops clever microchips that process very large data volumes in an energy-saving manner. This achievement impressed the jury of the pilot innovation competition “Energy-Efficient AI System” organized by the German Federal Ministry of Education and Research (BMBF). Here, Marco Breiling gives fascinating insights into his field of research.

Read the article

The SEC-Learn (Sensor Edge Cloud for Federated Learning) project, involving eleven Fraunhofer Institutes, promises a major technological leap in the field of neuromorphic hardware: for the first time, a chip is being developed to accelerate spiking neural networks (SNN) in conjunction with what is known as federated learning. This would offer practical benefits for companies and individuals.

Read the article

Bottlenecks are not exactly popular – after all, they bring entire processes to a standstill. Artificial intelligence has a bottleneck of its own to contend with. To be precise, it is impossible to speed up artificial intelligence indefinitely on classical computers. The hardware needs time to “think”.  Read here how we solve this problem.

Read the article

Computer architecture inspired by the human brain – to most people, this sounds rather abstract and removed from everyday life. Far from it: a research team from Fraunhofer IIS and Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU) has developed neuromorphic hardware capable of detecting early signs of atrial fibrillation and considerably reducing the risk of stroke. In other words, this application has an entirely practical place in everyday life.

Read the article

Bringing intelligence to devices is an essential prerequisite for autonomous driving. After all, real-life traffic situations demand fast responses and leave no time to transmit sensor data to a cloud for intelligent analysis. In the KI-FLEX project, researchers are developing neuromorphic hardware that allows artificial intelligence (AI) to be integrated directly into a car – in the form of a flexible platform. Project coordinator Michael Rothe from the Fraunhofer Institute for Integrated Circuits IIS offers his insights into the project.

To the article

Analog or digital – two terms that will be quite familiar to most people. But when it comes to artificial intelligence, only the experts know that there’s something “in between.” They call this “in between” the spiking neural network (SNN). One of the main hopes for the SNN approach is that it will resolve how to integrate “intelligence” directly into devices. Cognitive scientist Johannes Leugering explains what’s involved.

To the article

The purpose of Edge AI is to relocate artificial intelligence to where it is needed: end devices. For this to work, however, processing operations cannot be too energy-intensive. In the ECSEL projects ANDANTE and TEMPO, researchers at Fraunhofer IIS are paving the way for this emerging technology with the development of suitable chips, algorithms, and tools.

To the article

Some challenges are best faced shoulder to shoulder – like developing the chips, algorithms, and the hardware needed to integrate artificial intelligence directly into end devices, a paradigm also known as “Edge AI”. Marco Breiling, chief scientist at the Fraunhofer IIS Communication Systems division and Dr. Loreto Mateu, group manager at the Fraunhofer IIS Smart Sensing and Electronics division, talk about why their collaboration on Edge AI is so fruitful, and the benefits of working in an international team.

To the article