Speaker
Description
In this work, we present a detailed study of cross-talk in a linear single-photon avalanche
diode (SPAD) array using the LinoSPAD2 detector, which features 512 time-resolved
channels with a timing precision of 40 ps r.m.s. By characterizing and leveraging cross
talk effects, we were able to calibrate the intrinsic delays in the system readout, reducing
the uncertainty in the position of coincidence peaks from ±10 ns down to ±50 ps. While
such detectors are increasingly used in diverse fields – from quantum communications to
time-resolved imaging and quantum-assisted astronomy – their performance can be
critically affected by inter-channel cross-talk, especially in photon correlation
measurements such as the Hanbury Brown and Twiss (HBT) effect. Cross-talk can mimic
genuine correlation signatures, posing a fundamental challenge in experiments relying on
spatial or temporal intensity correlations.
We compare two versions of the LinoSPAD2 sensor: one equipped with microlenses to
enhance photon collection efficiency, and one without. Our measurements reveal a clear
difference in cross-talk behavior between the two. Specifically, we observe that the
addition of microlenses, while beneficial for photon detection efficiency, also leads to a
measurable increase in cross-talk probability between neighboring channels. We analyze
the spatial decay of cross-talk and discuss its implications for experiments sensitive to
second-order correlations. These findings offer valuable insight into the trade-offs
involved in SPAD array design and optimization for low-noise, high-resolution single
photon detection.
[1] Kulkov, Sergei, et al. "Inter-pixel cross-talk as background to two-photon interference
effects in SPAD arrays." Journal of Instrumentation 19.12 (2024): P12015.
[2] Kulkov, Sergei, et al. "Characterizing and exploiting cross-talk effect in SPAD arrays
for two-photon interference." arXiv preprint arXiv:2504.01185 (2025).
| Workshop topics | Detector systems |
|---|
