Learning diffractive optical communication around arbitrary opaque occlusions

• Published in: Nature communications Vol. 14; no. 1; p. 6830
• Main Authors: Rahman, Md Sadman Sakib, Gan, Tianyi, Deger, Emir Arda, Işıl, Çağatay, Jarrahi, Mona, Ozcan, Aydogan
• Format: Journal Article
• Language: English
• Published: London Nature Publishing Group UK 26-10-2023; Nature Publishing Group; Nature Portfolio
• Subjects: 639/624/1075; 639/624/1107/510; 639/624/399; Communication; Communications systems; Decoding; Deep learning; Encoders-Decoders; Free-space optical communication; Humanities and Social Sciences; Light speed; multidisciplinary; Neural networks; Occlusion; Science; Science (multidisciplinary); Space communication; Three dimensional printing; Transfer learning
• ISSN: 2041-1723; 2041-1723
• DOI: 10.1038/s41467-023-42556-0

Free-space optical communication becomes challenging when an occlusion blocks the light path. Here, we demonstrate a direct communication scheme, passing optical information around a fully opaque, arbitrarily shaped occlusion that partially or entirely occludes the transmitter’s field-of-view. In this scheme, an electronic neural network encoder and a passive, all-optical diffractive network-based decoder are jointly trained using deep learning to transfer the optical information of interest around the opaque occlusion of an arbitrary shape. Following its training, the encoder-decoder pair can communicate any arbitrary optical information around opaque occlusions, where the information decoding occurs at the speed of light propagation through passive light-matter interactions, with resilience against various unknown changes in the occlusion shape and size. We also validate this framework experimentally in the terahertz spectrum using a 3D-printed diffractive decoder. Scalable for operation in any wavelength regime, this scheme could be particularly useful in emerging high data-rate free-space communication systems.
Researchers demonstrate robust optical communication around fully opaque occlusions, partially or entirely blocking the light path, using a pair of electronic encoder and passive diffractive decoder that are jointly optimized using deep learning.