DE-TW-PNN
Photonic Edge AI Chips for Energy-Efficient High-Speed Applications
DE-TW-PNN develops photonic edge AI chips for energy-efficient high-speed applications. Photonic neural networks combine the processing speed of optical systems with the strengths of neural network architectures.
Objectives and Approach
The goal is to develop a photonic chip for AI data processing on edge devices, outperforming conventional electronic systems in energy efficiency and enabling AI training on edge hardware. The chip is manufactured in a 28-nm process.
Innovations and Perspectives
The project strengthens German-Taiwanese collaboration in microelectronics. The energy-efficient and high-performance photonic AI chip boosts innovation capacity at German research institutions and supports technological sovereignty goals.
Project Coordinator
- Technische Universität Berlin
Project Partners
- National Sun Yat-sen University, Kaohsiung
This project is funded by the German Federal Ministry for Research, Technology and Space (BMFTR) as part of the Design Initiative Microelectronics.
