Secure multi-party based cloud computing framework for statistical data analysis of encrypted data
MetadataVis full innførsel
Secure Multi-party Computation (SMC) is a paradigm used to accomplish a common computation among multiple users while keeping the data of each party secret from others. In recent years there has been a keen interest among the research community to look for techniques that can be adopted for the evolvement of SMC based solutions for improving its e ciency and performance. Cloud computing is a next generation computing solution in the eld of Information and Communication Technology (ICT) which allows its users to use high speed infrastructure and services provided by Cloud Service Providers (CSP) in a cost e ective manner with a higher availability. There- fore, deployment of cloud based architecture for SMCs would aid in improving its performance and e ciency. However, cloud based solutions raises concerns over secu- rity of users' private data, since data is handled by an external party that cannot be trusted. Hence, it is necessary to incorporate necessary security measures to ensure the security of users' private data. In this master's thesis we have addressed this issue by proposing a Secure Multi- party based Cloud Computing Framework which can ensure security, privacy and anonymity of users private data. In order to achieve this, we have formulated a case involving sales data analysis of a certain organization through computing statistical parameters of sales persons private sales data on a cloud environment. Furthermore, we have implemented a prototype of the proposed security framework which aids us to evaluate its performance. Moreover, considering the results that we have obtained, it is conclusive that cloud platforms can be successfully deployed to improve e ciency of SMCs while ensuring the security of users' private data; which in turn provides evidence for the practicability of multi-party based cloud computing solutions.
Masteroppgave i informasjons- og kommunikasjonsteknologi IKT590 2013 – Universitetet i Agder, Grimstad