Heisenberg-scaling characterization of an arbitrary two-channel network via two-port homodyne detection

Abstract

We present a fully Gaussian and experimentally feasible scheme for the simultaneous estimation of the four real parameters that characterize an arbitrary two-channel unitary transformation. The scheme utilizes a two-mode squeezed probe and balanced homodyne detection at both output ports, for which we derive the complete classical Fisher-information matrix analytically. Our scheme achieves the Heisenberg-scaling sensitivity for all four parameters simultaneously, enabling full multiparameter characterization of the generic two-channel interferometric device. We further show, by maximum-likelihood estimation, that the corresponding multiparameter Cramér-Rao bounds are saturated with a modest number of experimental repetitions and for low photon numbers. The scheme establishes a practical route to Heisenberg-scaling multiparameter Gaussian metrology for arbitrary two-channel networks, with direct relevance to calibration and sensing in integrated photonics and distributed quantum-enhanced measurement architectures.

Figures (6)

• Figure 1: A two-mode squeezed state (TMSS) (generated, for example, by an optical parametric amplifier (OPA)) probes an unknown two-channel linear-optical network described by an arbitrary unitary $U(\bm{\phi})$ in Eq. \ref{['Eq:Unitary']}, with parameter vector $\bm{\phi}=(\phi_0,\phi_1,\phi_2,\phi_3)$, where $\phi_3$ sets the mode mixing while $\phi_1,\phi_2$ and $\phi_0$ represent internal and overall phase parameters (the latter becomes operationally identifiable in the presence of a phase reference such as a local oscillator (LO)). The two output modes are measured via balanced homodyne detection (HD) at both ports, using tunable LO phases $\theta_1$ and $\theta_2$.
• Figure 2: The plot shows the Heisenberg-normalized scalar bound $N^{2}\,\mathrm{Tr}[F^{-1}]$ as a function of the total mean photon number $N$ for several choices of $k_1,k_2,k_3$. Since asymptotically $\mathrm{Tr}[F^{-1}] \propto 1/N^{2}$, the curves approach an $N$-independent plateau, which gives the leading prefactor set by the dominant $N^2$ contribution $\mathcal{F}$ to the Fisher information matrix in Eq. \ref{['eq:Ftot_asym_main']}. Varying $k_1,k_2,k_3$ therefore changes the prefactor while preserving the $1/N^{2}$ scaling. Here $\beta=1/2$.
• Figure 3: Maximum likelihood estimation of the four parameters under the two-port homodyne detection for $N=10$ and $\beta=1/2$. The main panel shows the normalized CRB $\Delta\widetilde{\phi}_i/(\Delta\phi_i)^{\mathrm{CRB}}$ with $i=0,1,2,3$ as a function of the number of measurement repetitions $M$, where $\Delta\widetilde{\phi}_i$ is the standard deviation of the estimator and $(\Delta\phi_i)^{\mathrm{CRB}}$ is the corresponding marginal CRB in Eqs. \ref{['eq:CRB_phi0']}--\ref{['eq:CRB_omega']}. The convergence of all four curves to unity confirms the saturation of the CRBs already for a relatively small number $M$ of experimental iterations, typically $M\approx 100$. In the inset, we plot the ratio between the expected value $\mathrm{E}[\widetilde{\phi}_i]$ of the maximum-likelihood estimator and the true value $\phi_i$. Here, $k_1=k_2=0.5$ and $k_3=0$.
• Figure 4: Maximum likelihood estimation of $\bm{\phi}$ as a function of the total mean photon number $N$ at a fixed sample size $M=200$. The main panel shows convergence of the normalized CRB $\Delta\widetilde{\phi}_i/(\Delta\phi_i)^{\mathrm{CRB}}$ for each parameter in Eqs. \ref{['eq:CRB_phi0']}--\ref{['eq:CRB_omega']} saturating the CRBs already at low photon numbers. The inset shows the unbiasedness ratio $\mathrm{E}[\widetilde{\phi}_i]/\phi_i$, confirming the estimator remains effectively unbiased for small values of $N$. Here, $k_1=k_2=0.5,k_3=0$ and $\beta=1/2$ for all $N$.
• Figure S1: Each panel shows the normalized effective Fisher information $1/(N^2\,(F^{-1})_{ii})$ plotted versus the total mean photon number $N$ for each of the four estimated parameters $\phi_0,\phi_1,\phi_2$, and $\phi_3$, respectively. Curves compare different LO detuning constants $k_1, k_2$ and balanced beam-splitter offset $k_3$. A plateau of $1/(N^2\,(F^{-1})_{ii})$ at large $N$ indicates that asymptotically $(F^{-1})_{ii}\propto 1/N^2$, i.e., a Heisenberg scaling of the marginal variance for that parameter. The vertical separation between curves reflects the prefactor of the asymptotic scaling: higher plateaus correspond to larger effective Fisher information and, therefore, higher achievable precision. Differences between panels highlight that the optimal working point need not be identical for all parameters. However the asymptotic Heisenberg scaling behavior holds for different values of $k_1,k_2,k_3$.