BIB-MKS: Post-Quantum Secure Biometric Identity-Based Multi-Keyword Search Over Encrypted Data in Cloud Storage Systems

Cloud computing technologies rely on powerful storage services and computing capabilities to maintain massive data for users. Since many cloud storage data probably contain sensitive privacy information, they are typically encrypted before outsourcing, which simultaneously limits the availability due to the difficulty of encrypted data retrieval. Public-key encryption with keyword search (PEKS) contributes to searching target encrypted data with keywords. However, majority of existing PEKS mechanisms require to manage certificates in the public key infrastructure (PKI), they are also vulnerable to adversaries equipped with quantum-computing devices in the near future. In this article, we devise a biometric identity-based multi-keyword search (BIB-MKS) mechanism from lattices over encrypted outsourced data, which inherently resists quantum-computing attacks. Each user in BIB-MKS is identified with her/his biometric information, which could be envisioned as the public key, thereby avoiding complex certificate managements. Particularly, BIB-MKS enables a data owner to produce an index associated with a biometric identity
$BID^{\prime }$
, such that a user with a biometric identity
$BID$
issues multiple keywords in a single search query and retrieves corresponding encrypted data, if and only if
$BID$
and
$BID^{\prime }$
are within a certain distance of each other as judged by some metric. BIB-MKS narrows down the search scope, and improves users search experience significantly. We define the formal security model of BIB-MKS, and prove the security of BIB-MKS under this model. The performance evaluation demonstrates that BIB-MKS is practical for post-quantum secure cloud storage systems.