Raphael Nägele from the Institute of Electrical and Optical Communications Engineering (INT) was awarded the “Gold Leaf Certificate” at the 20th International Conference on PhD Research in Microelectronics and Electronics (PRIME) 2025 in Taormina, Italy, for his paper "A 1.2-fJ/MAC, 7-fJ/ReLU, 5.9-b ENOB, All-Analog Neural Network Computing Macro."
The paper presents a highly energy-efficient, purely analog approach to AI accelerators—an important step toward more sustainable and powerful AI applications. Particularly noteworthy are the extremely low energy consumption per MAC operation (1.2 fJ) and ReLU activation (7 fJ) at an effective resolution of 5.9 bits.
The "Gold Leaf Certificate" is awarded to the top 10% of contributions.
The PRIME conference, one of the leading platforms for young scientists in microelectronics, offers ideal conditions for exchange with other international scientists and companies.
The article will be available shortly on our publications page.
Project: Fast and efficient analog processor for optical data links