SOA-based reservoir computing using up-sampling

TL;DR

This paper presents a novel SOA-based reservoir computing method that employs up-sampling and modulation, eliminating the need for delay loops, and demonstrates high prediction accuracy for complex time series.

Contribution

It introduces a delay-loop-free reservoir computing scheme using semiconductor optical amplifiers and photodetectors, enhancing simplicity and performance.

Findings

Achieved 400-step prediction of Mackey-Glass time series

Demonstrated effective nonlinear processing without delay loops

Validated the approach with experimental results

Abstract

We introduce a new approach to reservoir computing based on up-sampling and modulation, utilizing semiconductor optical amplifier and photodetector as nonlinear elements without conventionally used delay loop. We demonstrated the 400-step prediction capability of the proposed scheme for the Mackey-Glass time series test.