Right Device, Right Place: Variation-Aware Placement and Design Optimization for Robust Photonic Lattice-Filter Demultiplexers

2026-07-15

AI Summary

This work addresses the performance instability of integrated photonic lattice filter demultiplexers under fabrication errors and thermal perturbations by proposing a Bayesian co-optimization framework. For the first time, Bayesian optimization is applied to the joint optimization of device layout and design parameters, explicitly modeling process and thermal variations to enhance system robustness. By integrating variation-aware modeling with photonic device co-design, the proposed method achieves a 75% improvement in spectral matching performance and a 45% reduction in calibration power consumption in experimental validation, significantly outperforming existing approaches.

Abstract

We propose a Bayesian co-optimization framework for robust integrated photonic lattice-filter demultiplexers, jointly optimizing device placement and design parameters under fabrication and thermal variations. Results show 75% better spectral matching and 45% lower calibration power.

Problem

• photonic lattice-filter
• demultiplexers
• variation-aware placement
• design optimization
• fabrication variations

Innovation

• Bayesian co-optimization
• photonic lattice-filter
• variation-aware placement
• robust design
• demultiplexer