Identity-Based Encryption With Disjunctive, Conjunctive and Range Keyword Search From Lattices

To reduce data storage costs, more individuals are using cloud servers for reliable, scalable, cost-effective, and globally accessible solutions. However, storing data in plaintext on cloud servers can lead to data leakage risks. Moreover, the advancement of quantum computing poses a threat to traditional encryption algorithms. To counter quantum computing attacks and enable searches over encrypted keywords, lattice-based searchable encryption with conjunctive keyword search has been implemented. Nonetheless, existing schemes expose keyword fields and leaks additional information. To mitigate this, we propose a privacy-preserving method based on lattice hardness assumptions. It enables testing the existence of an encrypted keyword in a set of encrypted keywords without requiring the keyword fields. Additionally, we propose two improved methods: one for inclusion-based searches between two keyword sets, and another for range-based keyword searches. These form the basis for three lattice-based identity-based searchable encryption schemes that support disjunctive, conjunctive, and range keyword searches, respectively. The storage overhead of ciphertexts and trapdoors is unaffected by the number of keywords, making our scheme suitable for multi-keyword search scenarios. Our formal security analysis uses the learning with errors (LWE) assumption and our theoretical analysis and experimental simulations show comparable efficiency and low storage overhead.