Spiking Neural Networks – Fraunhofer IIS Researches AI Chip for Fiber Optic Networks
October 1, 2025 | FraunhoferEstimated reading time: 2 minutes
As the consortium leader of the EU-funded project SpikeHERO, Fraunhofer IIS is setting its sights on fiber optic networks over the next four years. Together with industry and research partners from the Netherlands, Czech Republic, and Belgium, the institute is developing AI chips based on spiking neural networks to significantly improve signal quality and boost data transmission rates in fiber optics. The project launches on October 1, 2025 and is intended to help expand Europe's digital infrastructure and consolidate its technological sovereignty.
Fiber optic networks are a cornerstone of the digital future. As industries compete for key technologies, demand for higher bandwidths and lower latencies continues to grow. Future 6G networks, in particular, must meet extremely high data transmission requirements. The challenge: the higher the desired data rate, the more the signal quality tends to degrade. To compensate for interference, providers currently rely on digital signal processors. “However, these come with high costs and are not a sustainable long-term solution due to their enormous power consumption,” explains Michael Rothe, manager of the Embedded AI group at Fraunhofer IIS. As a result, further increases in fiber optic data rates are hitting technical limits.
The SpikeHERO project is working on a solution. The project partners are developing a novel AI processor architecture that combines an optical and an electrical spiking neural network chip. These neural networks will continuously monitor the communication channel, analyze signals, and correct any interference at the receiver using control parameters. By maintaining the signal quality, new possibilities open up for increasing data rates in fiber optic systems. Specifically, the project aims to triple the bandwidth from 10 GHz to 30 GHz and reduce latency from 10 microseconds to under 6 nanoseconds. At the same time, energy consumption is expected to drop from 7–10 watts to just 1–2 watts.
Inspired by the brain
Spiking neural networks (SNNs) are considered a promising advancement in artificial intelligence. Their mode of operation mimics the principles of the human brain: information is processed in the form of pulses — or “spikes” — and only when a critical relevance threshold is exceeded. This makes SNNs ideal for AI applications that require both real-time responsiveness and energy efficiency.
There are different approaches to developing SNN chip hardware. Optical semiconductors transmit spikes via photons, while electrical counterparts use voltage and current. Each type has its advantages, and SpikeHERO aims to combine both. For the electrical SNN chip, the project will use the SENNA chip developed by Fraunhofer IIS and Fraunhofer EMFT. “We’re currently working on the second generation, which promises even higher spike rates with lower energy consumption,” says Rothe.
A pan-European collaboration
SpikeHERO (Spike Hybrid Edge Computing for Robust Optoelectrical Signal Processing) is funded by the European Innovation Council (EIC) with a budget of over €4.2 million and will run from October 1, 2025, to September 30, 2029. The initiative brings together research and industry partners from four European countries: Fraunhofer IIS, Fraunhofer EMFT, Eindhoven University of Technology, Hewlett Packard Enterprise Labs Belgium, and Argotech a.s. from the Czech Republic. Projects like SpikeHERO build on and strengthen the capabilities and infrastructure established through the Research Fab Microelectronics Germany (FMD). They also contribute to the APECS pilot line for Advanced Packaging and Heterogeneous Integration for Electronic Components and Systems, implemented by FMD as part of the EU Chips Act. The goal: to drive chiplet innovation and advance European semiconductor research and manufacturing in the long term.
Testimonial
"Advertising in PCB007 Magazine has been a great way to showcase our bare board testers to the right audience. The I-Connect007 team makes the process smooth and professional. We’re proud to be featured in such a trusted publication."
Klaus Koziol - atgSuggested Items
20 Years of Center Nanoelectronic Technologies (CNT) – Backbone of German Semiconductor Research Celebrates Anniversary
08/14/2025 | Fraunhofer IPMSThe Center Nanoelectronic Technologies (CNT) of the Fraunhofer Institute for Photonic Microsystems (IPMS) is celebrating its 20th anniversary this year. Since its founding in 2005, it has developed into a pillar of applied semiconductor research in Germany and Europe. With its unique research cleanroom and equipment adhering to the 300-mm wafer industry standard, CNT is unparalleled in Germany and serves as a central innovation driver for the microelectronics industry.
Fraunhofer IPMS, DIVE Optimize Semiconductor Processes with Cutting-Edge Measurement System
07/17/2025 | Fraunhofer IPMSThe Fraunhofer Institute for Photonic Microsystems IPMS, in collaboration with DIVE imaging systems GmbH, has achieved a major milestone in resource-efficient semiconductor manufacturing.
NY CREATES, Fraunhofer Institutes Announce Joint Development Agreement to Advance Memory Devices at the 300mm Wafer Scale
05/16/2025 | NY CREATESNY CREATES and Fraunhofer IPMS announced at a signing ceremony a new Joint Development Agreement (JDA) to drive research and development focused on memory devices.
Smart and Compact Sensors with Edge-AI
04/16/2025 | FraunhoferA newly launched interdisciplinary research project involving universities of Brandenburg and research institutions is developing new technological approaches for better and more effective integration of artificial intelligence at the edges of IT networks, so-called “edges”.
Miniaturized ISFET pH Sensor Optimized for Easy Use
04/10/2025 | FraunhoferThe Fraunhofer Institute for Photonic Microsystems IPMS has achieved another milestone in chemical liquid analysis. The electronics required to control the ion-sensitive field effect transistors (ISFET) have been successfully miniaturized.