Lightweight Auditable Multi-Keyword Attribute-Based Searchable Encryption Scheme for Secure Health Data Sharing

Attribute-based searchable encryption (ABSE) scheme combines attribute-based encryption with searchable encryption, which enables fine-grained access control during ciphertext retrieval, effectively meeting privacy concerns of medical data sharing in cloud environments. However, existing ABSE schemes suffer from high computational overhead due to numerous pairing operations during encryption and decryption process, which makes them impractical for resource-constrained scenarios such as the Internet of Things (IoT). Furthermore, existing schemes also lack effective supervision of post-authorization data sharing processes, resulting in potential misuse of medical data by authorized users. To address the above limitations, we propose a lightweight auditable multi-keyword attribute-based searchable encryption scheme (LA-MABSE) for secure health data sharing. The encryption phase of the scheme is divided into offline and online phases: computationally intensive tasks are preprocessed offline, while the online phase only performs lightweight operations, which greatly reduces the computational overhead of the user-side encryption. Meanwhile, the outsourcing decryption mechanism is introduced to delegate partial decryption tasks to edge servers, further reducing the computational overhead in the decryption phase. To enhance accountability, we introduce the supervisory chain to record data access behaviors and a dynamic membership management algorithm is also designed to revoke the attributes related to malicious users who misuse of data. Finally, we analyze the security of the scheme, and the theoretical analysis and experimental results show that the scheme has better efficiency performance.