Application Of Privacy-Preserving Private Information Retrieval For The Cloud Environment

With the rapid development of the cloud computing technology, more and more organizations and individuals outsource their data storage needs into cloud and using cloud-based Information technology resources provide data management, information processing, sharing and exchange. However, because of the complexity of the cloud computing environment and the uncertainty of the user behavior, the client privacy can’t be protected well. It is important to provide some methods to ensure not only cooperative bilateral to sharing of information need, but also the private information is not leaked to adversary. Private Information Retrieval (PIR) was proposed to resolve such a problem:when clients submit queries to the data storage side, the intension of these queries should be hidden. For the existing problems in Private Information Retrieval research such as high complexity, low efficiency of protocol performance, short of extended Implementation in the cloud environment, lack of server data privacy protection, poor practicability, etc. focusing on the PIR protocol, protocol parallel implementation, communication complexity optimization, the practicality of public key encryption based keyword search protocol using PIR, etc. we conduct deep research on three aspects.In order to lower the computation complexity of PIR protocols, A Computational PIR Protocol called hcPIRMR based on additively homomorphic encryption scheme and MapReduce was proposed, which is implemented using MapReduce computing framework for cloud environments. In the stage of query processing, hcPIRMR distribute outsourcing data storage to different working nodes in cloud environment, using "Parallel" and "aggregation" phase of MapReduce computing framework, the original computational overhead being amortized through sharing to multiple working nodes’ computing power in a query jobs, thus improving the overall efficiency of the protocol execution.In user multi-query scenario, batch queries and query cache communication complexity optimization methods is proposed. General method in the execution of PIR protocol, a query vector same size of file quantity is construct and submit to data storage side, and another query need to rebuild the query vector again to submit the server, which resulting in a large number of the user and server-side communication overhead. In multiple queries, we construct batch queries method based on query elements offset, through caching the initial query, subsequent query only transfer elements offset based on the initial query elements, The method effectively reduce the amount of communication between the two sides.In order to solve PIR protocol’ practicability and inadequate server side’ data privacy protection issue, public key encryption based keyword search using PIR scheme is presented, which combined public key encryption system with private information retrieval, and which use proposed protocol hcPIRMR as the building-block as well as perfect hash function tool to construct a cloud keyword-based encrypted data retrieval scheme, apply to cloud environment outsourcing scenarios. The scheme combine protect access patterns of PIR protocol with the effectiveness of public key encryption and security of outsourcing data. The solution finish the cloud encryption data keyword-based queries, assure query efficiency and the cloud data security.