Hardware implementation of artificial neural networks facilitates real-time parallel processing of massive data sets. Optical neural networks offer low-volume 3D connectivity together with large bandwidth and minimal heat production in contrast to electronic implementation. Here, we present a DMD based approaches to realize energetically efficient light coupling into a multi-core fiber realizing a unique design for in-fiber optical neural networks. Neurons and synapses are realized as individual cores in a multi-core fiber. Optical signals are transferred transversely between cores by means of optical coupling. Pump driven amplification in Erbium-doped cores mimics synaptic interactions. In order to dynamically and efficiently couple light into the multi-core fiber a DMD based micro mirror device is used to perform proper beam shaping operation. The beam shaping reshapes the light into a large set of points in space matching the positions of the required cores in the entrance plane to the multi-core fiber.
Eyal Cohen, Dror Malka, Amir Shemer, Asaf Shahmoon, Michael London, and Zeev Zalevsky, "Micro mirrors based coupling of light to multi-core fiber realizing in-fiber photonic neural network processor," Proc. SPIE 10117, Emerging Digital Micromirror Device Based Systems and Applications IX, 1011703 (Presented at SPIE OPTO: January 30, 2017; Published: 20 February 2017); https://doi.org/10.1117/12.2253426.
Conference Presentations are recordings of oral presentations given at SPIE conferences and published as part of the conference proceedings. They include the speaker's narration along with a video recording of the presentation slides and animations. Many conference presentations also include full-text papers. Search and browse our growing collection of more than 14,000 conference presentations, including many plenary and keynote presentations.