Quantum computing holds the promise of providing an exponential speedup for several computational tasks. Quantum repeaters could allow long-distance entanglement generation which can, in turn, enable distributed quantum computation, secure communication, and precision sensing. However, building a useful quantum computer or quantum repeater using currently known architectures is beyond current experimental capabilities. Photon-mediated quantum information processing may be a path to realizing such devices because of the scalability offered by recent advances in integrated photonics and the natural role of photons as information carriers. Furthermore, photonic qubits do not suffer from decoherence, unlike ion-trap and superconducting qubits. On the other hand, photonic architectures must often contend with other non-idealities like photon loss and the probabilistic nature of linear optics. The resource requirements for building an all-optical quantum repeater capable of beating the repeaterless bound, using multiplexing based creation of photonic cluster states, are studied. We find several improvements which reduce the resource requirements by five orders of magnitude. We then analyze a ``one-way" repeater based on the quantum parity code which reduces the resource requirements by another order of magnitude. In order to further reduce the resource requirements, ideas from percolation theory can be used to create resource states for universal quantum computing from 3-photon GHZ states without feed-forward. We develop a new framework for studying such percolation-based creation of photonic clusters which is used to find better lattices and find the the limits of such an approach. We use a similar idea to develop an architecture for cluster state creation in a system of atomic memories connected via photonic links, with an analysis focussed on nitrogen vacancy (NV) centers in diamond. Finally, we develop an entanglement routing protocols for quantum networks in which every node only needs to perform entanglement swaps.
Thesis Supervisor: Prof. Dirk Englund