Neuromorphic Computing with Nonlinear Dynamics in Photonic Crystal Fiber

Neuromorphic computing based on photonic nonlinear dynamics is gaining popularity, due to its low power consumption, as well as their scalability [1]. However, how nonlinearity scales with the network performance are rarely investigated. Here, we numerically investigate the performance of neuromorphic computing, using photonic crystal fiber as a reservoir media [2]. Photonic crystal fiber has a high nonlinearity and thus, capable to produce spectrum that ranges from linear propagation, up to octave spanning supercontinuum generation.