Multi-parameter two-photon polarimetry at the quantum limit

Technical Summary: Multi-Parameter Two-Photon Polarimetry at the Quantum Limit

Problem Statement
Photonic quantum metrology has established advantages in precision and resource efficiency for single-parameter estimation, often approaching the fundamental quantum Cramér-Rao bound (QCRB). However, extending these advantages to the simultaneous estimation of multiple parameters is hindered by the intrinsic incompatibility of quantum measurements. Generally, the measurements required to saturate the QCRB for individual parameters are mutually incompatible, and uncertainty relations prevent the simultaneous achievement of maximum precision for all parameters. While schemes exist for multi-parameter estimation, they are often limited to small regions of the parameter space or require adaptive strategies to extend coverage. Furthermore, practical implementations must account for experimental imperfections, such as imperfect state preparation and reduced interference visibility, which are often omitted in theoretical proposals.

Methodology
The authors present an experimental protocol for the simultaneous estimation of two polarization parameters, $\theta$ and $\delta\phi$, encoded on a two-photon state $|\psi\rangle = |\alpha_1\rangle|\alpha_2\rangle$. The single-photon states are defined as $|\alpha_i\rangle = \cos(\theta/2)|H\rangle + \sin(\theta/2)e^{i\phi_i}|V\rangle$, where $\theta$ represents the shared polar angle and $\delta\phi = |\phi_1 - \phi_2|/2$ represents half the difference in azimuthal angles.

The experimental setup utilizes a two-photon interferometer. Photon pairs are generated via Type-II parametric down-conversion in an aperiodically poled KTP crystal pumped by a Ti:sapphire laser. The photons are prepared in specific polarization states using wave plates and then interfere at a balanced beam splitter (BS). The measurement scheme involves projecting the output photons onto horizontal and vertical polarization bases using polarizing beam splitters (PBS) and detecting them with superconducting nanowire single-photon detectors (SNSPD). To handle experimental noise and distinguishability, the authors introduce an indistinguishability parameter $\eta$ and derive a modified Fisher information matrix $F'$. Maximum likelihood estimators for $\tilde{\theta}$ and $\tilde{\delta\phi}$ are derived, accounting for $\eta$ and the specific detection outcomes (coincidence in same/different ports and polarization combinations).

Key Contributions

1. Robust Theoretical Framework: The authors extend the theoretical framework proposed in Ref. [32] to include experimental imperfections, specifically imperfect polarization states and reduced interference visibility ($\eta < 1$). They demonstrate that while the precision for $\theta$ remains unaffected by $\eta$, the precision for $\delta\phi$ is bounded by $\eta^2$, with the maximum saturation of the QCRB occurring at $\delta\phi = \pi/4$.
2. Experimental Realization: The team implements the scheme across a wide range of the parameter space, operating in both the asymptotic (many-sample) and non-asymptotic (few-sample) regimes.
3. Few-Photon Efficiency: The protocol achieves near-optimal performance with as few as $\sim 200$ photon pairs, making it suitable for applications involving dim sources or photosensitive samples where minimizing optical exposure is critical.

Results

• Asymptotic Regime: Using $N=2000$ samples, the estimators showed good agreement with nominal state preparations. The bias for $\tilde{\theta}$ was approximately $0.03$ rad across the range. For $\tilde{\delta\phi}$, the bias was within $0.03$ rad in the central region ($\delta\phi \in [\pi/4, 3\pi/4]$) but increased to an average of $0.13$ rad for extremal values, attributed to imperfect state preparation and the sensitivity of the protocol to small differences in polar angles ($\delta\theta$) at these points.
• Few-Sample Regime: The authors evaluated the bias and variance for sample sizes $N \in \{40, 120, 200\}$. The estimator $\tilde{\theta}$ was found to be unbiased around $\theta = \pi/2$, with the variance quickly saturating the classical Cramér-Rao bound (CRB) as $N$ increased. For $\tilde{\delta\phi}$, the unbiased region and saturation region were wider for $\theta = \pi/2$ compared to $\theta = \pi/4$.
• QCRB Saturation: For $N=200$, the protocol achieved a saturation of the QCRB of $(93 \pm 2)\%$ for $\tilde{\theta}$. For $\tilde{\delta\phi}$, the saturation was $(89 \pm 2)\%$ at $\theta = \pi/2$ and $(86 \pm 3)\%$ at $\theta = \pi/4$. The authors note that the saturation for $\tilde{\delta\phi}$ is fundamentally bounded by the indistinguishability factor $\eta^2 \approx 95\%$.

Significance and Claims
The paper claims to demonstrate a photonic multi-parameter estimation scheme that approaches the ultimate quantum Cramér-Rao precision bound simultaneously for two polarization parameters. The significance lies in the ability to achieve this high precision across a wide parameter space using a minimal number of resources ($\sim 200$ photon pairs). The authors highlight that this approach offers a distinct advantage over standard quantum state tomography (QST), which would require up to three times as many photon pairs to achieve comparable precision due to the need for sequential measurements in multiple bases. By utilizing interference to estimate both parameters simultaneously within a single polarization basis, the protocol offers a more resource-efficient pathway for polarimetric sensing, particularly for applications involving fragile samples or low-light conditions such as X-ray astronomy. The work acknowledges that the degree of saturation is limited by state preparation imperfections, an unavoidable consequence of measurement incompatibility in multi-parameter estimation, but asserts that the scheme remains robust and effective within these physical constraints.