Privacy protection in outsourced database services

YongHong Yu,Wenyang Bai

DOI: https://doi.org/10.1109/MEC.2011.6025814

2011-01-01

Abstract:Technical considerations and many significant commercial and legal regulations demand that privacy guarantees be provided whenever sensitive information is stored, processed, or communicated to external partied. In this paper, we propose a solution to enforce data confidentiality, data privacy and accountable user privacy in outsourced database services. The approach starts from a flexible definition of privacy constraints, applies encryption on information in a parsimonious way and mostly relies on attribute partition to protect sensitive information. Based on the approximation algorithm for the minimal encryption attribute partition, the approach allows storing the outsourced data on un-trusted database server and minimizing the amount of encrypted data. By combining cryptographic with auxiliary random server, the approach can reduce the computational and communication complexity of private information retrieval to provide user privacy protection. By introducing verifiable encryption and revocation of decryption based on event capsule, the approach obtains accountability when the user misbehaves. The theoretical analysis shows that our new approach can provide efficient data privacy, efficient accountable user privacy protection with lower computational complexity and not increase the cost of communication complexity simultaneously. © 2011 IEEE.