Probabilistic Cutoffs in Homogeneous Quantum Repeater Chains

Quantum repeater chains will use heralded entanglement generation and entanglement swaps to create entangled links between distant end-nodes in a quantum network. However, entangled links inside the repeater chain decohere while waiting to be swapped. This can lead to a low fidelity of the end-to-end link. Previous work has considered time-dependent cutoff policies that provide strict control of the fidelity of all links in the chain by limiting their storage time. Assuming entanglement is generated in discrete time steps, we introduce a new type of cutoff policy in which a link is discarded with fixed probability at the end of each time step. The resulting probabilistic cutoff policy does not require tracking and communicating the storage times of entangled links, but this simplicity comes at the cost of abandoning strict control of the fidelity. We study the effect of this on repeater chain performance by comparing the probabilistic cutoff policy to different time-dependent cutoff policies. We find that the probabilistic cutoff policy can retain a significant fraction of the secret-key rate that can be achieved with a time-dependent cutoff policy, especially in short chains. We also identify performance advantages of the probabilistic cutoff policy over the time-dependent cutoff policies, most notably when memory coherence times are short.