Performance analysis of near-term quantum networks

Abstract

Quantum networks hold the potential to enable new applications, such as secure key distribution, high-precision distributed sensing, and distributed quantum computing. A central functionality of a quantum network is the distribution of entanglement between remote parties. Since experimental implementations remain in an early stage, it is important to understand both the capabilities and limitations of near-term architectures. However, characterising quantum network performance is challenging, due to the complex, stochastic nature of even simple architectures. Analytical studies can therefore play a crucial role: they not only reduce computational cost but also reveal fundamental relationships between performance, the choice of entanglement distribution protocols, and properties of quantum network hardware. In this thesis, we develop analytical methods to study quantumnetwork performance in several important scenarios.

We firstly analyse entanglement buffers, which are systems designed to generate and store high-quality entangled states to be consumed at any time. For this setting, we derive analytical expressions for two key performance metrics. The solutions are computationally efficient, make no restrictive assumptions about the entanglement purification protocol, and allow general insights: for example, that simple purification schemes can outperformmore complex ones previously considered “optimal” in different contexts.

Then,we turn to the problemof entanglement packet generation,where multiple entangled states of sufficient quality must be established simultaneously between network users. The fast generation of entanglement packets is an essential capability for many quantum network protocols. We obtain analytical results for the entanglement packet generation rate under a constant entanglement generation scheme and later extend the analysis to adaptive schemes, where entanglement parameters are tuned dynamically. Using parameter regimes motivated by current experiments, we show that adaptivity can enhance the entanglement packet generation rate by up to a factor of twenty.

Finally, we examine a standard assumption in performance analyses: that the initial states in a quantum repeater chain can be approximated by a symmetrised, or “twirled”, form. We investigate this assumption in the contexts of postselected and non-postselected entanglement swapping, where postselection is performed based on the Bell-state measurement outcomes at the repeaters. A central result is that, in many relevant cases, the twirled approximation is exact for non-postselected swapping. More generally, we provide a systematic framework to determine when the twirled approximation is valid for the initial states of a repeater chain.