Computational Scaling in Inverse Photonic Design Through Factorization Caching

TL;DR

This paper introduces a factorization caching method that significantly accelerates inverse photonic design simulations by reusing linear system solutions, enabling faster optimization of complex nanophotonic devices.

Contribution

The authors develop a caching approach for factorized matrices in FDFD simulations, drastically reducing computation time in inverse photonic design processes.

Findings

Achieved up to 9.2x speedup in device simulations.

Scales efficiently across device sizes from 16μm² to 7000μm².

Enhances computational efficiency for complex photonic device design.

Abstract

Inverse design coupled with adjoint optimization is a powerful method to design on-chip nanophotonic devices with multi-wavelength and multi-mode optical functionalities. Although only two simulations are required in each iteration of this optimization process, these simulations still make up the vast majority of the necessary computations, and render the design of complex devices with large footprints computationally infeasible. Here, we introduce a multi-faceted factorization caching approach to drastically simplify the underlying computations in finite-difference frequency-domain (FDFD) simulations, and significantly reduce the time required for device optimization. Specifically, we cache the symbolic and numerical factorizations for the solution of the corresponding system of linear equations in discretized FDFD simulations, and re-use them throughout the entire device design process. As proof-of-concept demonstrations of the resulting computational advantage, we present simulation speedups reaching as high as $9.2\times$ in the design of broadband wavelength and mode multiplexers compared to conventional FDFD methods. We also show that factorization caching scales well over a broad range of footprints independent of the device geometry, from as small as $16{\mu m}^2$ to over $7000 {\mu m}^2$. Our results present significant enhancements in the computational efficiency of inverse photonic design, and can greatly accelerate the use of machine-optimized devices in future photonic systems.