Combining a passive spatial photonic reservoir computer with a semiconductor laser increases its nonlinear computational capacity

Abstract

Photonic reservoir computing has been used to efficiently solve difficult and timeconsuming problems. The physical implementations of such reservoirs offer low power consumption and fast processing speed due to their photonic nature. In this paper, we investigate the computational capacity of a passive spatially distributed reservoir computing system. It consists of a network of waveguides connected via optical splitters and combiners. A limitation of its reservoir is that it is fully linear and that the nonlinearity - which is often required for solving computing tasks - is only introduced in the output layer. To address this issue, we investigate the incorporation of an additional active nonlinear component into the system. Our approach involves the integration of a single semiconductor laser in an external optical delay line within the architecture. Based on numerical simulations, we show that the architecture with this semiconductor laser has a nonlinear computational capacity that is significantly increased as compared to the original passive architecture, which can be beneficial to solving difficult computational tasks.