Secure Multi-Party Sorting and Applications

摘要: Research within the area of cryptography constitutes core this the- sis. In addition to cryptography, we also present results in peer-assisted streaming and web security. We on two specific cryptographic problems: broadcast encryption secure multi-party computation. Broad- cast is problem efficiently securely distributing content a large changing group receivers. Secure computation subject how number parties can collaborate securely. All all, thesis spans from systems work discussing Spotify system with millions users, more theoretic, foundational results. Streaming among largest applications Internet today. On- demand services allow users consume media they want, at their convenience. With catalogs offered by many services, access wide selection content. Live provides means for corporations as well individuals world. The power such broadcasts was shown recent (early 2011) revolts Tunisia Egypt, where protesters streamed live demonstrations. To stream global audience requires significant resources, particular terms bandwidth needed. One approach reduce requirements use peer-to-peer techniques, clients assist media. commercial music-on-demand system, using streaming. thesis, discuss protocol measurements its performance. systems, it important restrict solutions encryption. Within area, first which improves efficiency previous cost lowered secu- rity guarantees. second contains lower-bound proofs, showing that early protocols subset cover framework are essentially optimal. Many web-based, user accesses browser. Apart usage web, subscriptions bought This provide service, must understand result new type attack, an old history detection vulnerability time execution redirect victim’s computation, has three contributions. Firstly, give efficient basic functions summation disjunction adapt network run on. Secondly, sorting aggregation, techniques networks. Finally, prove dichotomy theorem, all distinct outputs either maximally easy or difficult regards security provided.