Self-Configuring and Reconfigurable Silicon Photonic Signal Processor

Photonic signal processing is essential in the optical communication and\noptical computing. Numerous photonic signal processors have been proposed, but\nmost of them exhibit limited reconfigurability and automaticity. A feature of\nfully automatic implementation and intelligent response is highly desirable for\nthe multipurpose photonic signal processors. Here, we report and experimentally\ndemonstrate a fully self-learning and reconfigurable photonic signal processor\nbased on an optical neural network chip. The proposed photonic signal processor\nis capable of performing various functions including multichannel optical\nswitching, optical multiple-input-multiple-output descrambler and tunable\noptical filter. All the functions are achieved by complete self-learning. Our\ndemonstration suggests great potential for chip-scale fully programmable\noptical signal processing with artificial intelligence.\n