Towards realistic architectures for linear optical quantum computing

  • Sam Morley-Short

Student thesis: Doctoral ThesisDoctor of Philosophy (PhD)

Abstract

With the end of Moore’s Law within sight, quantum computers offer a tantalising paradigm shift in computational power. Currently, many quanta are competing to realise such a revolutionary device, of which this thesis considers one in particular: linear optical quantum computation (LOQC). Over the past decade LOQC architectures have developed from “efficient” but unfeasible toy models to serious contenders. A significant step in previous works was the blueprint of an LOQC architecture that could be conceivably implemented with idealised optical components.
However, in reality nature is not kind and devices not ideal. As such, we consider open questions addressing gaps between LOQC’s theoretical architecture and experimental constraints. In doing so, a selection of numerical tools are also developed for the design, simulation and analyses of novel architectures. Specifically, we consider three problems.
Firstly, can an infinite-sized quantum state be realised within a finite-sized device? Through development of a simple, generalised model, we find some small, finite device size at which the infinite state is faithfully reproduced. We also find that increasing device size above this confers no advantage, thereby identifying some necessary and sufficient minimum LOQC device size.
Secondly, we consider the challenge of accommodating unheralded photon loss in an LOQC architecture, a problem for which no previous solution was known. By developing a novel protocol for optimal teleportation on stabilizer states, we show that unheralded loss may be tolerated, perhaps entirely, by adaptive measurement strategies.
Finally, we consider the optimisation of LOQC architectures via local complementation. This work both sets hard limits on the states accessible by postselected linear optics circuits as well as develops novel tools for the analysis of higher-dimensional quantum states.
We conclude with an example of how such works can be combined to optimise the LOQC architecture as well as provide improved device resource estimates.
Date of Award19 Mar 2019
Original languageEnglish
Awarding Institution
  • The University of Bristol
SupervisorJohn G Rarity (Supervisor) & Hugo Cable (Supervisor)

Keywords

  • quantum computing
  • linear optical quantum computing
  • photonics
  • quantum computing architectures

Cite this

'