Solitonic waveguide reflection at an electric interface

Episodic Memory and Information Recognition Using Solitonic Neural Networks Based on Photorefractive Plasticity

Neuromorphic models are proving capable of performing complex machine learning tasks, overcoming the structural limitations imposed by software algorithms and electronic architectures. Recently, both supervised and unsupervised learnings were obtained in photonic neurons by means of spatial-soliton-waveguide X-junctions. This paper investigates the behavior of networks based on these solitonic neurons, which are capable of performing complex tasks such as bit-to-bit information memorization and recognition. By exploiting photorefractive nonlinearity as if it were a biological neuroplasticity, the network modifies and adapts to the incoming signals, memorizing and recognizing them (photorefractive plasticity). The information processing and storage result in a plastic modification of the network interconnections. Theoretical description and numerical simulation of solitonic networks are reported and applied to the processing of 4-bit information.

Addressable Refraction and Curved Soliton Waveguides Using Electric Interfaces

A great deal of interest over the years has been directed to the optical space solitons for the possibility of realizing 3D waveguides with very low propagation losses. A great limitation in their use for writing complex circuits is represented by the impossibility of making curved structures. In the past, solitons in nematic liquid crystals, called nematicons, were reflected on electrical interfaces, and more recently photorefractive spatial solitons have been, as well. In the present work, we investigate refraction and total reflection of spatial solitons with saturable electro-optic nonlinearity, such as the photorefractive ones, on an electric wall acting as a reflector. Using a custom FDTD code, the propagation of a self-confined beam was analyzed as a function of the applied electric bias. The electrical reflector was simulated by applying different biases in two adjacent volumes. We observed both smaller and larger angles of refraction, up to the critical π/2-refraction condition, and then the total reflection. The radii of curvature of the associated guides can be varied from centimeters down to hundreds of microns. The straight guides showed losses as low as 0.07 dB/cm as previously observed, while the losses associated with single curves were estimated to be as low as 0.2 dB.