FPGA-Based Adaptive Control for Phase Stabilization in Fiber-Optic Interferometers Using Correlated Photons

TL;DR

This work addresses phase noise in time-bin quantum interferometry by implementing an FPGA-based adaptive Perturbation-and-Observe controller driven solely by coincidence counts. By dynamically adjusting the perturbation step within a circular constraint, the method achieves up to ~70% faster rise-time and substantial noise suppression, with visibility improvements sustained for over 600 seconds. The approach is model-free, hardware-efficient, and scalable, enabling robust long-term phase stabilization in quantum communication systems. The results demonstrate significant practical impact for deploying stable, high-contrast quantum interference in fielded fiber-optic networks.

Abstract

Time-bin encoded photon pairs enable robust, decoherence-resistant transmission through optical fibers for long-distance quantum communication, where phase noise poses a critical limitation to stable operation. Here, we implement an adaptive Perturbation-and-Observe algorithm on a fully digital FPGA platform operating with real-time feedback at 1 Hz. The control signal is derived from the coincidence counts of correlated photon pairs. This adaptive approach reduces the rise time by 70\% and the coincidence noise by 30\%, resulting in visibility improvements sustained for more than 600 s.These results provide an efficient solution for long-term phase stabilization in quantum and photonic systems.

Figures (6)

• Figure 1: Optical setup to generate single photons in time-bin state superpositions. Correlated photon pairs are generated by Spontaneous parametric down-conversion, being one of them detected as a trigger (Tg) in a heralded single-photon source scheme. The other goes through two Franson interferometers (unbalanced Mach-Zenhder interferometer) in sequence with fiber stretchers (ST1 and ST2) in their long arms modulating the longitudinal phase.
• Figure 2: FPGA-based phase acquisition and control. Detector signals yield singles (CS) and coincidences (CC), driving an APC module with P&O and adaptive P&O. Control outputs (via DAC and AMP) actuate phase stretchers (ST1/2), closing the loop. The CS and CC are also read by an embedded CPU and streamed to a host PC. Solid arrows denote feedback of control paths, and dashed arrows denote readout paths.
• Figure 3: Classical Perturb-and-Observe (P&O) implemented in FPGA. Inputs: Current intensity $I_\text{actual}$, previous intensity $I_\text{prev}$, fixed step size $p$, enable signal $EnControl$, synchronization pulse $sync$. Output: Control signal $S_{ctrl}$.
• Figure 4: Sawtooth signal used to self-adjust the circular constraint. Small amplitude yields incomplete modulation (a). Optimal amplitude ensures continuous cosine–sinusoidal output (b). Large amplitude causes discontinuities (c).
• Figure 5: Operation of the control system showing coincidence counts for controlled and uncontrolled intervals (top), and the performance of the conventional (a) and adaptive (b) P&O algorithms at the stretcher output (bottom).