Prediction of intermittent chaos in a semiconductor laser with optical feedback using reservoir computing

Abstract Prediction of intermittent chaos is an important issue when detecting the precursors of extreme events such as catastrophic phenomena. We numerically investigate the time-series prediction of intermittent chaos to detect and predict the occurrence of extreme events in a semiconductor laser using reservoir computing (RC). Intermittent chaos is observed in a semiconductor laser subjected to optical self-feedback, and irregular switching occurs suddenly between burst and laminar states. We perform k -step-ahead prediction and investigate the possibilities for the prediction of the occurrence of chaotic bursts using RC.