Photonic reservoir computer based on frequency multiplexing

TL;DR

This paper presents a photonic reservoir computer utilizing frequency multiplexing to encode neuron states, enabling high-speed processing of multiple channels for tasks like channel equalization and time series forecasting.

Contribution

It introduces a novel photonic reservoir computing system that exploits frequency domain multiplexing for neuron encoding and optical implementation of output weights.

Findings

Processes 25 neurons simultaneously at 20 MHz

Achieves good performance on benchmark tasks

Demonstrates optical implementation of output weights

Abstract

Reservoir computing is a brain inspired approach for information processing, well suited to analogue implementations. We report a photonic implementation of a reservoir computer that exploits frequency domain multiplexing to encode neuron states. The system processes 25 comb lines simultaneously (i.e. 25 neurons), at a rate of 20 MHz. We illustrate performances on two standard benchmark tasks: channel equalization and time series forecasting. We also demonstrate that frequency multiplexing allows output weights to be implemented in the optical domain, through optical attenuation. We discuss the perspectives for high speed high performance low footprint implementations.