A double selection entanglement distillation-based state estimator

1. Problem Statement

As quantum communication networks evolve toward a future "Quantum Internet," there is a critical need for efficient protocols to characterize and certify quantum resources (specifically entangled Bell pairs) without incurring excessive resource overhead.

• Current Limitations: Traditional characterization methods like full tomography or direct fidelity estimation are resource-intensive and often require separate protocols that interrupt network operations.
• Existing Solutions: A previous method, the "Disti-Mator" [15], integrated estimation into entanglement distillation. However, it required running three distinct distillation protocols to characterize the Bell-diagonal elements of the input states. This reconfiguration is impractical for users needing a consistent supply of high-fidelity states or strict timing coordination.
• The Gap: There is a need for a more compact estimator that utilizes a single distillation protocol to estimate both the noisy input states and the resulting distilled states, thereby minimizing setup changes and resource consumption.

2. Methodology

The authors propose a novel estimator based on the Double Selection Distillation Protocol [16].

The Protocol

The protocol involves two parties (Alice and Bob) sharing three identical noisy Bell pairs ($\rho^{(0)}, \rho^{(1)}, \rho^{(2)}$):

1. Bilateral CNOTs: Alice and Bob perform CNOT gates between $\rho^{(0)}$ (control) and $\rho^{(1)}$ (target), followed by CNOTs between $\rho^{(1)}$ (control) and $\rho^{(2)}$ (target).
2. Measurements: They measure $\rho^{(1)}$ in the $Z$ basis and $\rho^{(2)}$ in the $X$ basis.
3. Post-selection: They communicate results classically. If coincidences are confirmed in both bases, $\rho^{(0)}$ is successfully distilled; otherwise, it is discarded.

Estimation Strategy

The input state is modeled as a Bell-diagonal state $\rho(q)$ defined by convex coefficients $q = (q_1, q_2, q_3, q_4)$. The authors assume the input states are independent and identical, and that the distillation hardware is perfect (noise only exists in the source).

The core innovation is mapping the measurement statistics of the protocol to the unknown parameters:

1. Transformation: The authors define intermediate variables $x_i$ related to the coefficients $q$.
2. Three Equations: They derive three coincidence probabilities from the protocol:
   • $p(Z)$: Probability of coincidence in the $Z$ basis.
   • $p(X)$: Probability of coincidence in the $X$ basis.
   • $p(XZ)$: Joint probability of coincidence in both bases (signaling successful distillation).
3. Inversion: They prove that these three probabilities uniquely determine the variables $x_1, x_2, x_3$ via invertible functions (some involving cube roots and square roots). These $x$ values are then transformed back to estimate the Bell-diagonal coefficients $\hat{q}$.

Theoretical Guarantees

• Concentration Bounds: Using Hoeffding's inequality, the authors derive rigorous bounds on the estimation error. They show that the trace distance between the true state and the estimated state is bounded with high probability, depending on the number of trials ($N$).
• Distilled State Estimation: The method allows for "free" estimation of the distilled state by propagating the error bounds of the input state through the known quantum instrument (the distillation map).

3. Key Contributions

1. Single-Protocol Estimator (Double Selection Disti-Mator): The paper demonstrates that a single double-selection protocol is sufficient to estimate all Bell-diagonal parameters of the input state, eliminating the need for the three-protocol setup required by previous methods.
2. Analytical Inversion: The authors provide explicit analytical formulas to invert the measurement statistics ($p(Z), p(X), p(XZ)$) into state parameters ($q$), proving the system is solvable.
3. Resource Efficiency: They prove that this new estimator is more resource-efficient (in terms of sample complexity) than the original Disti-Mator [15] for a wide range of high-fidelity states.
4. Dual Utility: The protocol serves a dual purpose: it acts as a state estimator while simultaneously performing entanglement distillation, allowing for the recovery of high-fidelity states during the estimation process.

4. Results

The authors validated their approach through numerical simulations and theoretical analysis:

• Comparison with Tomography:
  • For input states with high fidelity ($q_1$ close to 1), the Double Selection Disti-Mator requires a comparable number of resources to standard single-copy tomography.
  • For low-fidelity states, tomography is more efficient, but the Disti-Mator offers the unique advantage of potentially yielding a distilled state, whereas tomography consumes all resources for measurement only.
• Comparison with Original Disti-Mator [15]:
  • The new estimator significantly outperforms the original Disti-Mator in sample complexity for a large region of parameter space (specifically for high-fidelity inputs).
  • It avoids the overhead of reconfiguring the setup for three different protocols.
• Feasibility Tests:
  • Simulations using realistic parameters (e.g., $q = (0.62, 0.15, 0.05, 0.18)$ and $N=10^4$ triplets) showed a mean trace distance error of $\approx 0.010$.
  • For higher fidelity inputs ($q_1 = 0.88$), the error dropped to $\approx 0.003$, demonstrating that the estimator is highly accurate for near-term quantum network conditions.
  • The distribution of errors was unimodal and positively skewed, indicating reliable performance.

5. Significance

This work addresses a critical bottleneck in the deployment of quantum networks: resource-efficient certification.

• Operational Efficiency: By integrating estimation directly into the distillation task, network operators can monitor resource quality in real-time without pausing operations to run separate benchmarking protocols.
• Scalability: The reduction in sample complexity and the elimination of multi-protocol reconfiguration make this approach more scalable for large-scale quantum networks.
• Future Directions: The paper opens the door for designing estimators for other complex distillation protocols, moving toward a paradigm where characterization is a natural byproduct of network operations rather than a separate cost.

In summary, the authors have successfully developed a compact, mathematically rigorous, and resource-efficient method to characterize quantum states using the inherent statistics of a double-selection distillation protocol, offering a practical solution for the emerging quantum internet.