Evaluation of Distimation's Real-world Performance on a Superconducting Quantum Computer

TL;DR

This paper presents Distimation, a distillation-based protocol for efficient Bell-diagonal state estimation in quantum networks, addressing the scalability limitations of Quantum State Tomography. By extracting Bell-diagonal parameters from distillation success probabilities, the method enables near real-time entanglement monitoring, validated via simulations and IBM superconducting hardware. The study shows strong performance in controlled Pauli-noise simulations but reveals limitations under real hardware noise and finite-shot budgets; it also demonstrates robustness in asymmetric, MBQC-based network scenarios. The results highlight the practical potential of Distimation for real-world quantum networks and point to future work in robust post-processing, dynamic noise modeling, and extensions to multi-qubit entanglement and fault-tolerant contexts.

Abstract

Quantum state estimation plays a crucial role in ensuring reliable creation of entanglement within quantum networks, yet conventional Quantum State Tomography (QST) methods remain resource-intensive and impractical for scaling. To address these limitations, we experimentally validate Distimation, a novel distillation-based protocol designed for efficient Bell-diagonal state estimation. Using IBM Quantum simulators and hardware, we demonstrate that Distimation accurately estimates Bell parameters under simulated and real-world noise conditions, but also demonstrating limitations with operational noise and number of available shots. Additionally, we simulate an asymmetric-fidelity Bell pair scenario via Measurement-Based Quantum Computation (MBQC) to further validate Distimation under realistic network conditions. Our results establish Distimation as a viable method for scalable, real-time entanglement monitoring in practical quantum networks.

Figures (10)

• Figure 1: Quantum circuits for entanglement purification: distillation-(a)(left) detects X and Y errors; distillation-(b)(middle) detects Z and Y errors; and distillation-(c)(right) detects states with X and Z errors. Each circuit begins with two entangled pairs and applies distinct purification logic.
• Figure 2: Qubit layout of ibm_kawasaki, obtained from IBM Quantum Platform.
• Figure 3: Trace distance $D(\hat{\rho}_W, \rho_W)$ across Werner parameter space under simulated Pauli noise. Minimal variation from zero demonstrates that our distimation of Werner states functions correctly.
• Figure 4: Trace distance $D(\hat{\rho}_{\text{BD}}, \rho_{\text{BD}})$ across Bell-diagonal parameter space under simulated Pauli noise. Bell-diagonal states cannot be correctly estimated with Werner Distimation procedures alone but require the full Bell-diagonal Distimation.
• Figure 5: Quantum circuit for Quantum State Tomography with swap.