LAUSR

In parallel with the rapid developments of cloud-assisted IoT, the corresponding security and privacy issue emerges as a challenge. Biometric recognition technologies are interesting and promising to reinforce traditional cryptographic and personal authentication systems for cloud-assisted IoT. However, the biometric information, if compromised, cannot be canceled and substituted easily. In this paper, based on fuzzy commitment protocol, a fingerprint recognition scheme using minutiae-based sector coding strategy is proposed for cloud-assisted IoT. In our approach, the minutiae of a fingerprint are classified into many designed sectors and then encoded according to their features. With the idea of fuzzy commitment, the key encryption process is accomplished by using BCH codes and Hash mappings. Our scheme, not only provides a flexible way to realize the fingerprint recognition between the complexity and security by designing error-correcting codes with different parameters, but also offers a good balance between the genuine accept rate (GAR) and the false acceptance rate (FAR) with customized sector coding strategies.