Emulation of Entanglement Distribution Networks on a Quantum Computer

A comparative study of depolarizing noise implementations for teleportation-based multipartite entanglement distribution

Ashley N. Tittelbaugh; Jerry Horgan; Rohan Bali; Marco Ruffini; Daniel C. Kilper; Shelbi L. Jenkins; Boulat A. Bash· 2026· DOI 10.48550/arXiv.2607.14260

The core problem

Quantum networks promise secure communication, distributed quantum computing, and enhanced sensing, but their performance is limited by noise and latency. Emulating these networks on quantum computers offers a controllable testbed to study such impairments. This paper focuses on teleportation-based distributed multipartite-entanglement-state construction, a key primitive for quantum network protocols. The authors model imperfect Bell-pair sources using depolarizing noise channels and classical communication delays using thermal relaxation. They implement depolarization via three methods: Stinespring dilation, randomly applied Pauli errors, and quasi-probability decompositions. The latter two are evaluated on IQM quantum hardware, while all three are simulated. The central question is how these mathematically equivalent noise models perform under realistic hardware constraints.

Innovation

The authors find that although the three depolarizing noise models are mathematically equivalent, their hardware implementations yield significantly different results. On IQM quantum hardware, the random Pauli error method and quasi-probability decomposition produce diverging fidelities for the distributed entangled state. Specifically, the quasi-probability method, which involves negative coefficients and sampling, shows higher variance and lower average fidelity due to hardware noise and limited sampling. The random Pauli method, while simpler, also deviates from the ideal simulation. In contrast, the Stinespring dilation, only simulated, provides a baseline that matches theoretical predictions. The thermal relaxation model for latency further degrades fidelity, with longer delays leading to greater decoherence. The results highlight that hardware constraints—such as gate errors, readout errors, and connectivity—cause the mathematically equivalent models to behave differently in practice.

Quantum networks promise secure communication, distributed quantum computing, and enhanced sensing, but their performance is limited by noise and latency. Emulating these networks on quantum computers offers a controllable testbed to study such impairments. This paper focuses on teleportation-based distributed multipartite-entanglement-state construction, a key primitive for quantum network protocols. The authors model imperfect Bell-pair sources using depolarizing noise channels and classical communication delays using thermal relaxation. They implement depolarization via three methods: Stinespring dilation, randomly applied Pauli errors, and quasi-probability decompositions. The latter two are evaluated on IQM quantum hardware, while all three are simulated. The central question is how these mathematically equivalent noise models perform under realistic hardware constraints.

The authors consider a quantum network where entanglement is distributed via Bell pairs subject to depolarizing noise. The depolarizing channel is defined as:

Why it matters

The profound differences observed underscore the importance of careful experiment design when emulating quantum networks on quantum computers. The choice of noise implementation is not merely a technical detail; it affects the conclusions drawn about network performance. The quasi-probability decomposition, while theoretically elegant, is particularly sensitive to hardware noise due to its use of negative coefficients and post-processing. The random Pauli method is more robust but still limited by gate errors. The Stinespring dilation, though ideal for simulation, may be challenging to implement on hardware due to the need for controlled environment interactions. The authors suggest that future work should focus on developing noise models that are both mathematically equivalent and hardware-efficient. They also note that the thermal relaxation model for latency is a simplification; more accurate models should include both amplitude and phase damping. Overall, this study provides a cautionary tale: emulating quantum networks requires not only correct mathematics but also a deep understanding of the underlying hardware.

Who should read this

CS practitioners and researchers

Opening member content…