A Novel Fuzzy Identity-Based Encryption with Wireless PUF for IoT Device Authentication.

Identity-Based Encryption algorithm generates public-private key with identity, reduces the key management complexity of IoT devices. However, traditional IBE algorithm is vulnerable to spoofing attacks. This paper proposes a novel Fuzzy Identity-Based Encryption algorithm, defines the fuzzy identity as an attribute set composed of multidimensional physical features of IoT devices. The algorithm extracts device fuzzy identity from their unique physical characteristics, establish a linkage between device physical attributes and logical identity. Moreover, this paper introduces the concept of fuzziness to overcome the impact of physical noise. We calculate the fuzziness of individual attributes within the attribute set, as well as the overall fuzziness among the sets to match fuzzy identity. To achieve low-cost device identity authentication for multiple wireless IoT devices in a contactless scenario, this paper proposes a Physical Unclonable Function to extract device-related physical attributes based on Channel State Information. The stimuli response of the PUF constitute the fuzzy identity of IoT devices. Formal security analysis indicates that the security of our method can be reduced to the hardness of the Decisional Modified Bilinear Diffie-Hellman (MBDH) assumption and provides Chosen Ciphertext Attack security. Our approach achieves a 96.9% success rate in extracting public-private keys despite physical noise, even in scenarios with concurrent access by multiple wireless IoT devices.