Comparative Robustness of Mach-Zehender and Microring Photonic Matrix Multipliers Under Laser RIN and Thermal Drift
Abstract
Photonic matrix multipliers can accelerate neural-network inference, but their practical utility depends on how optical weight elements respond to source noise, thermal variation, and calibration error. This paper compares Mach-Zehnder modulator (MZM) and microring-resonator (MRR) implementations under laser relative-intensity noise (RIN), column-wise thermal/phase drift, and their joint action. Two CIFAR-10 classifiers - a compact four-layer CNN and ResNet-8 - are mapped to matched photonic multiply-accumulate chains using common source-distribution, detection, numerical, and calibration assumptions. Both architectures reproduce the same zero-perturbation baseline for each workload: 79.5% classification accuracy for the Small CNN and 70.0% for ResNet-8. Robustness is then evaluated using separate RIN-only and drift-only sweeps and a full two-dimensional RIN-drift grid, with dimensionless RMS perturbation levels extending from 0.5% to 20%. The analysis reports classification accuracy, class normalized confusion matrices, rolling-accuracy traces, one-dimensional robustness curves, and multidimensional response surfaces. Device equations and parameter definitions follow the models implemented in the QPI Labs Photonic Inference Simulator, with shared laser fluctuations propagated as correlated multiplicative noise and slow drift represented as a per-output-column perturbation of calibrated weights. All numerical results are generated from recorded experiment configurations and random seeds. The study provides a controlled system-level framework for comparing interferometric and resonant photonic weight elements while limiting conclusions to the implemented device models, perturbation ranges, calibration policy, and neural-network workloads.
Related
- OptiLookUp: A Microring-Based Photonic Read-Only Memory Lookup Table Engine for Photonic Accelerators
- On-chip high-repetition-rate passively mode-locked frequency comb source based on a microring composite cavity
- Self-Pulsing Microring Resonator Networks for Bandwidth-Efficient Event Detection in an Optical Fiber Sensor
- Fading Memory and Nonlinear Dynamics in a Microring Resonator Network and their Use in Computation
- Co-design of Silicon Microring Modulator beyond 200 Gb/s per Lane: Device Physics, Operating Point, and Compact Models for Scale-Up and Scale-Out Optical I/O
BibTeX
@misc{falkenberg2026comparative,
title = {{Comparative Robustness of Mach-Zehender and Microring Photonic Matrix Multipliers Under Laser RIN and Thermal Drift}},
author = {Andreas Falkenberg},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.15400},
eprint = {2610.15400},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.15400},
}
RIS
TY - UNPB
TI - Comparative Robustness of Mach-Zehender and Microring Photonic Matrix Multipliers Under Laser RIN and Thermal Drift
AU - Andreas Falkenberg
PY - 2026
DA - 2026/10/11
PB - arXiv
DO - 10.48550/arxiv.2610.15400
UR - https://arxiv.org/abs/2610.15400
AB - Photonic matrix multipliers can accelerate neural-network inference, but their practical utility depends on how optical weight elements respond to source noise, thermal variation, and calibration error. This paper compares Mach-Zehnder modulator (MZM) and microring-resonator (MRR) implementations under laser relative-intensity noise (RIN), column-wise thermal/phase drift, and their joint action. Two CIFAR-10 classifiers - a compact four-layer CNN and ResNet-8 - are mapped to matched photonic multiply-accumulate chains using common source-distribution, detection, numerical, and calibration assumptions. Both architectures reproduce the same zero-perturbation baseline for each workload: 79.5% classification accuracy for the Small CNN and 70.0% for ResNet-8. Robustness is then evaluated using separate RIN-only and drift-only sweeps and a full two-dimensional RIN-drift grid, with dimensionless RMS perturbation levels extending from 0.5% to 20%. The analysis reports classification accuracy, class normalized confusion matrices, rolling-accuracy traces, one-dimensional robustness curves, and multidimensional response surfaces. Device equations and parameter definitions follow the models implemented in the QPI Labs Photonic Inference Simulator, with shared laser fluctuations propagated as correlated multiplicative noise and slow drift represented as a per-output-column perturbation of calibrated weights. All numerical results are generated from recorded experiment configurations and random seeds. The study provides a controlled system-level framework for comparing interferometric and resonant photonic weight elements while limiting conclusions to the implemented device models, perturbation ranges, calibration policy, and neural-network workloads.
ER -