Nyquist-Sampled Time-Domain Adjoint FDTD for Memory-Efficient Broadband Nanophotonic Inverse Design
Abstract
Adjoint optimization is a cornerstone of broadband nanophotonic inverse design, but conventional time-domain implementations face a severe memory bottleneck because they retain forward-field histories at every finite-difference time-domain (FDTD) time step. Here, we show that this full time-step storage is unnecessary for broadband design objectives because the underlying fields are band-limited. By storing forward fields only at Nyquist intervals and using the resulting sparse fields during the adjoint pass, the proposed method enables on-the-fly gradient accumulation without retaining full forward-field histories. This Nyquist-sampled adjoint FDTD framework preserves the two-simulation scaling of time-domain adjoint optimization while substantially reducing the dominant field-storage. Because the broadband gradient is evaluated directly in the time domain, with no spectral discretization, the per-iteration cost is independent of the number of frequencies---in contrast to frequency-sampled adjoint formulations, whose cost grows with spectral sampling density. Gradient verification confirms that Nyquist sampling reproduces conventional full-storage adjoint gradients with negligible error, whereas undersampling beyond the Nyquist limit produces aliasing-induced gradient degradation. Across four two-dimensional broadband nanophotonic benchmarks and a fully three-dimensional metalens, the method maintains gradient fidelity and optimized device performance while reducing dominant field-storage memory by more than 100× relative to full-history storage in a prototypical example. These results suggest that the principal memory barrier in broadband time-domain adjoint FDTD is not an intrinsic requirement of gradient evaluation but rather a consequence of redundant temporal field storage, thereby opening a practical route to large-scale three-dimensional nanophotonic inverse design.
Related
- Ultra-broadband (1.3–2.5 μm) and highly efficient graphene electro-optic modulator based on a buried silicon waveguide
- Limitations of Nyquist Criteria in the Discretization of 2-D Electromagnetic Integral Equations at High Frequency: Spectral Insights Into Pollution Effects
- Identification and characterization of 15 265 super-Nyquist frequencies in 1309 δ Scuti stars from Kepler photometry
- Unified light-matter metric of molecular and nanophotonic chirality
- Versatile Sleeve-and-Bulk Fabrication of Inversely Designed Nanophotonic Structures with High Pattern Transfer Fidelity
BibTeX
@misc{park2026nyquistsampled,
title = {{Nyquist-Sampled Time-Domain Adjoint FDTD for Memory-Efficient Broadband Nanophotonic Inverse Design}},
author = {Mingyu Park and Owen D. Miller and Haejun Chung},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2607.08159},
eprint = {2607.08159},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2607.08159},
}
RIS
TY - UNPB
TI - Nyquist-Sampled Time-Domain Adjoint FDTD for Memory-Efficient Broadband Nanophotonic Inverse Design
AU - Mingyu Park
AU - Owen D. Miller
AU - Haejun Chung
PY - 2026
DA - 2026/09/08
PB - arXiv
DO - 10.48550/arxiv.2607.08159
UR - https://arxiv.org/abs/2607.08159
AB - Adjoint optimization is a cornerstone of broadband nanophotonic inverse design, but conventional time-domain implementations face a severe memory bottleneck because they retain forward-field histories at every finite-difference time-domain (FDTD) time step. Here, we show that this full time-step storage is unnecessary for broadband design objectives because the underlying fields are band-limited. By storing forward fields only at Nyquist intervals and using the resulting sparse fields during the adjoint pass, the proposed method enables on-the-fly gradient accumulation without retaining full forward-field histories. This Nyquist-sampled adjoint FDTD framework preserves the two-simulation scaling of time-domain adjoint optimization while substantially reducing the dominant field-storage. Because the broadband gradient is evaluated directly in the time domain, with no spectral discretization, the per-iteration cost is independent of the number of frequencies---in contrast to frequency-sampled adjoint formulations, whose cost grows with spectral sampling density. Gradient verification confirms that Nyquist sampling reproduces conventional full-storage adjoint gradients with negligible error, whereas undersampling beyond the Nyquist limit produces aliasing-induced gradient degradation. Across four two-dimensional broadband nanophotonic benchmarks and a fully three-dimensional metalens, the method maintains gradient fidelity and optimized device performance while reducing dominant field-storage memory by more than 100× relative to full-history storage in a prototypical example. These results suggest that the principal memory barrier in broadband time-domain adjoint FDTD is not an intrinsic requirement of gradient evaluation but rather a consequence of redundant temporal field storage, thereby opening a practical route to large-scale three-dimensional nanophotonic inverse design.
ER -