Hardware and Protocol Optimization in Quantum-Repeater Networks

The future quantum internet promises to enable users all around the world to, among other applications, generate shared secure keys and perform distributed quantum computations. To do so, entanglement must be distributed between remote users. One way of doing this is by sending photons through optical fiber, which allows for reusing some existent classical infrastructure. However, the probability of photons being absorbed in optical fiber grows exponentially with the distance covered, rendering entanglement generation at larger-than-metropolitan scales unfeasible. One possible approach to enable distributing entanglement over larger distances is to employ quantum repeaters, devices that can in theory mitigate the effects of fiber loss by splitting the total distance to be covered into smaller segments. Despite recent advances, the required technology is still under development. In this dissertation we aim to contribute to a swifter realization of fiber-based quantum-repeater networks.

To this end, we introduce a methodology combining quantum-network simulations and genetic-algorithm-based optimizations that allows for determining hardware requirements for quantum repeaters. Using this methodology we translate quantum-network-application-derived performance metrics into specific requirements on the quantum repeaters used to implement the quantum network. This indicates not only how good hardware must be in order to enable given applications, but also in what specific ways state-of-the-art hardware must be improved to do so.

We also investigate the effects of using existing fiber infrastructure for the deployment of near-term quantum networks. Doing so would be a cost-effective way of constructing quantum networks. However, existing infrastructure also imposes constraints, namely on where quantum hardware can be placed. We quantify to what extent such constraints affect quantum-network performance, as well as how these effects can be mitigated by optimizing repeater placement.

Finally, we contribute to answering the question of how to extract the best possible performance out of imperfect hardware. For a given hardware quality, making the right choices with regards to what protocols are executed by the nodes and where nodes are placed can result in significant boosts in performance. We perform a joint hardware-protocol optimization and find that good hardware choices can significantly relax hardware requirements, as well as highlight multiple possible paths to functional quantum-repeater networks. We also provide tools for the discovery of entanglement generation protocols.