Ultrafast perturbation maps as a quantitative tool for testing of multi-port photonic devices

Abstract

Advanced photonic probing techniques are of great importance for the development of non-contact wafer-scale testing of photonic chips. Ultrafast photomodulation has been identified as a powerful new tool capable of remotely mapping photonic devices through a scanning perturbation. Here, we develop photomodulation maps into a quantitative technique through a general and rigorous method based on Lorentz reciprocity that allows the prediction of transmittance perturbation maps for arbitrary linear photonic systems with great accuracy and minimal computational cost. Excellent agreement is obtained between predicted and experimental maps of various optical multimode-interference devices, thereby allowing direct comparison of a device under test with a physical model of an ideal design structure. In addition to constituting a promising route for optical testing in photonics manufacturing, ultrafast perturbation mapping may be used for design optimization of photonic structures with reconfigurable functionalities.

Deep learning enabled design of complex transmission matrices for universal optical components

“…For the initial pattern generation (step 0), we use an inhouse implementation of the aperiodic Fourier Modal Method (a-FMM) 37,38 . The a-FMM relies on a supercell method and perfectly matched layers to compute the modes of the system, and a scattering matrix formalism to describe mode coupling and has previously been shown to provide numerical predictions with high accuracy and fast convergence 39 .…”

“…The a-FMM relies on a supercell method and perfectly matched layers to compute the modes of the system, and a scattering matrix formalism to describe mode coupling and has previously been shown to provide numerical predictions with high accuracy and fast convergence 39 . It has also been demonstrated to be well-suited to the study of multiport MMI devices containing perturbations 19,38 , owing to its very fast runtime and direct access to the full set of coupling coefficients between all input and output waveguide modes. Simulations are performed in 2D using the effective index method 40 .…”

Deep Learning Enabled Design of Complex Transmission Matrices for Universal Optical Components

“…Pale and dark colors correspond to initial and final network iterations, respectively. the transmittance is nearly unity 47 . This can be seen in the qualitative comparison to the electric field simulation shown in Fig.…”

“…In principle any Maxwell solver can be used for the evaluation of the fitness function f . For the initial pattern generation (step 0), we use an in-house implementation of the aperiodic Fourier Modal Method (a-FMM) 46,47 , relying on a supercell method and perfectly matched layers to compute the modes of the system, and a scattering matrix formalism to describe mode coupling. The method has previously been shown to provide numerical predictions with high accuracy and fast convergence 48 .…”

“…The method has previously been shown to provide numerical predictions with high accuracy and fast convergence 48 . It has also been demonstrated to be well-suited to the study of multi-port MMI devices containing perturbations 19,47 , thanks to its very fast runtime and direct access to the full set of coupling coefficients between all input and output waveguide modes. Simulations are performed in 2D using the effective index method 49 .…”