Companion Preprint: Dynamic Link-Loss / Polarization-Dependent Efficiency in Photonic Bell Tests
1 view
Skip to first unread message
Christopher ONeil
Jun 1, 2026, 7:19:54 PM (6 hours ago) Jun 1
to Bell inequalities and quantum foundations
Hi everyone,
Following up on the preprint of the Propagation Delay and Temporal Coincidence Bounds manuscript that I shared a few days ago, I wanted to circulate the companion paper: "Uncharacterized In-Situ Polarization-Dependent Efficiency and Dynamic Eberhard Bounds in Photonic Bell Tests".
These two manuscripts are directly complementary, representing two physical pillars of the same underlying metrological phenomenon. Both analyze the physical transient consequences of high-speed nanosecond EOM switching in loophole-free Bell test architectures, but they explore completely different physical domains:
1. The Temporal Pillar (Propagation Delay and Temporal Coincidence Bounds): Analyzes the temporal consequence. Active EOM transitions draw massive capacitive current ($2.0\text{ A}$), inducing localized EMI and ground sags on comparator rails. Coupled with dynamic PMD, this shifts photon arrival time tags by $10\text{--}30\text{ ps}$. In tight-coincidence postselected pipelines (such as high-rate DI-QKD), this acts as a temporal postselection filter, shifting correlations by $\Delta S \approx 0.001\text{--}0.005$.
2. The Spatial-Polarization Pillar (Polarization-Dependent Efficiency and Dynamic Eberhard Bounds):Analyzes the spatial and polarization consequence. EOM driver voltage ringing drives rapid spatial beam walk-off ($5.5\,\mu\text{m}$) across the active nanowire meanders and uncompensated polarization rotation ($13.5^{\circ} - 18^{\circ}$). This creates a setting-dependent link-loss / coupling attenuation ($\eta_{\text{var}}$). Under Emmerson's total-variation selection framework, we prove this reweights the classical integration space, yielding an upper classical CHSH limit of $S \approx 2.264$ (closing ~32% of the classical-to-quantum gap).
Together, these papers make a rigorous, unified case for why static, common-mode hardware calibrations are structurally insufficient for dynamic loophole-free Bell tests and DI-QKD security architectures.
While the Propagation Delay manuscript shows how transients modulate the coincidence filter, the Polarization-Dependent Efficiency manuscript shows how they modulate the detection efficiency itself. Both outline non-circular, time-resolved in-situ calibration protocols to empirically isolate and bound these hardware non-idealities under operational loads.
The code and schematic assets for the Polarization-Dependent Efficiency model are available in the updated code repository at: https://github.com/chris-oneil/photonic-bell-test-transients
I would love to get your thoughts.
Warm Regards,
Christopher O'Neil
Following up on the preprint of the Propagation Delay and Temporal Coincidence Bounds manuscript that I shared a few days ago, I wanted to circulate the companion paper: "Uncharacterized In-Situ Polarization-Dependent Efficiency and Dynamic Eberhard Bounds in Photonic Bell Tests".
These two manuscripts are directly complementary, representing two physical pillars of the same underlying metrological phenomenon. Both analyze the physical transient consequences of high-speed nanosecond EOM switching in loophole-free Bell test architectures, but they explore completely different physical domains:
1. The Temporal Pillar (Propagation Delay and Temporal Coincidence Bounds): Analyzes the temporal consequence. Active EOM transitions draw massive capacitive current ($2.0\text{ A}$), inducing localized EMI and ground sags on comparator rails. Coupled with dynamic PMD, this shifts photon arrival time tags by $10\text{--}30\text{ ps}$. In tight-coincidence postselected pipelines (such as high-rate DI-QKD), this acts as a temporal postselection filter, shifting correlations by $\Delta S \approx 0.001\text{--}0.005$.
2. The Spatial-Polarization Pillar (Polarization-Dependent Efficiency and Dynamic Eberhard Bounds):Analyzes the spatial and polarization consequence. EOM driver voltage ringing drives rapid spatial beam walk-off ($5.5\,\mu\text{m}$) across the active nanowire meanders and uncompensated polarization rotation ($13.5^{\circ} - 18^{\circ}$). This creates a setting-dependent link-loss / coupling attenuation ($\eta_{\text{var}}$). Under Emmerson's total-variation selection framework, we prove this reweights the classical integration space, yielding an upper classical CHSH limit of $S \approx 2.264$ (closing ~32% of the classical-to-quantum gap).
Together, these papers make a rigorous, unified case for why static, common-mode hardware calibrations are structurally insufficient for dynamic loophole-free Bell tests and DI-QKD security architectures.
While the Propagation Delay manuscript shows how transients modulate the coincidence filter, the Polarization-Dependent Efficiency manuscript shows how they modulate the detection efficiency itself. Both outline non-circular, time-resolved in-situ calibration protocols to empirically isolate and bound these hardware non-idealities under operational loads.
The code and schematic assets for the Polarization-Dependent Efficiency model are available in the updated code repository at: https://github.com/chris-oneil/photonic-bell-test-transients
I would love to get your thoughts.
Warm Regards,
Christopher O'Neil
Independent Researcher
Big Rapids, MI, USA
Reply all
Reply to author
Forward
0 new messages