The problem of secure computation considers a set of parties who do not trust each other and yet want to perform public computations on data sets held privately by each individual. The most important property of secure computations is that they are input-privacy preserving. Intuitively this means that after the computation has finished no party can say something meaningful about the inputs of the other parties -beyond what is implicit in the outcome of the computation. This thesis focuses on the design and analysis of protocols in the pre-processing model with a majority of dishonest parties. This model presupposes an initial set-up that produces a large number of independent data to be used as auxiliary input at a later stage, when the actual secure computation takes place. The' current state of the art for producing such precomputed data employs homomorphic encryption and techniques derived from oblivious transfer. As a side effect we also dwell on the problem of extending oblivious transfer.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:689689 |
| Date | January 2015 |
| Creators | Vega, Enrique Larraia de |
| Publisher | University of Bristol |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
Page generated in 0.0021 seconds