A Traceable CP-ABE Scheme Supporting Dynamic Revocation and Efficient Decryption for Medical Data Sharing
Hongjian Yin, Jiaxuan Li, Yan Zhu, Lei Zhang, Baojun Qiao · IEEE Internet of Things Journal · 2025
Ciphertext-Policy Attribute-Based Encryption (CP-ABE) is suited for securing electronic health records (EHRs), as it enables fine-grained access control over sensitive medical data while preserving interoperability. However, existing schemes face significant challenges in dynamic user revocation and efficient decryption, including privacy leakage during revocation and difficulties in tracing malicious users. To address these issues, we propose T-DRED, a traceable and dynamic revocation-enabled decryption scheme based on secure aggregation functions. T-DRED employs a secure aggregation function to dynamically insert revoked user identities into a revocation list, enabling selective ciphertext updates and reducing revocation overhead. To preserve privacy, revocation status is verified through membership determination without revealing user identities. To improve decryption efficiency, our scheme innovatively adopts a two-phase decryption process, separating user identity verification from policy verification. In the first phase, lightweight membership verification identifies valid users; in the second phase, attribute-policy matching is performed. Additionally, each user key contains a white-box tracing identifier, allowing the system to trace and revoke malicious users upon key leakage. Revoked users are denied access to both new and historical ciphertexts via list updates, realizing integrated trace-and-revoke functionality. In terms of security, under the standard model and based on theq-BDHE andl-SDH assumptions, the IND-CPA, traceability, and membership verification security of our T-DRED scheme are formally proven. Experimental results show that T-DRED improves encryption and decryption efficiency over existing related schemes.