Abstract Objectives: To develop a trusted key support model for synchronized protection-key derivation that improves duplicate-aware cloud storage by reducing dependence on direct content-only key generation without sacrificing computational efficiency. Methods: The proposed model derives a 128-bit synchronized protection key from transformed user data and a trusted support key supplied by a separate service component. The derivation process includes block conversion, bit-count-based transformation, XOR-based reduction, trusted support-key integration, and final key generation. The model was implemented as a cloud application using C#.NET on Microsoft Azure and evaluated against HMAC and PMAC using execution time for file sizes of 10 KB, 20 KB, 30 KB, 40 KB, 50 KB, and 100 KB. Findings: The proposed method consistently outperformed the baseline techniques across all tested input sizes. At 10 KB, the execution time was 6 ms, compared with 12 ms for HMAC and 10 ms for PMAC. At 50 KB, the proposed method required 32 ms, whereas HMAC and PMAC required 59 ms and 50 ms, respectively. At 100 KB, the proposed method completed in 63 ms, compared with 123 ms for HMAC and 102 ms for PMAC. The percentage improvement ranged from 45.76% to 51.35% over HMAC and from 36.00% to 43.90% over PMAC. These results show that trusted support can be incorporated into synchronized protection-key derivation while maintaining duplicate-aware storage behavior and achieving lower computational overhead. Novelty: The study introduces a trusted-service-assisted synchronized protection-key derivation model for duplicate-aware cloud storage and demonstrates, through implementation-based evaluation, that the method provides consistent runtime improvement over HMAC and PMAC while reducing reliance on direct content-only derivation. Keywords: Cloud storage, Data deduplication, Trusted key support, Synchronized data protection, Key derivation
G Uma (Wed,) studied this question.