.png){kind=icon}
.png){kind=icon}
Return
Secret Key Generation by Continuous Encryption Before Quantization
.png){kind=icon}
OA
AI
DOI：10.1109/LSP.2022.3184253.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Secret key generation (SKG) is a fundamental signal processing problem for security applications including wireless network security and biometric template security. For SKG, a pair (or more) of highly correlated secret vectors (SVs) need to be respectively quantized into a pair (or more) of almost identical sequences of binary bits (i.e., keys). The literatures on SKG in the wireless community almost uniformly default on a direct quantization on SVs. On the other hand, many works on biometric template security advocate a quantization on the output of a one-way function of SVs. In this paper, we present a generalized approach for SKG called continuous encryption before quantization (CEbQ). By CEbQ, a pair of SVs of limited dimension are first transformed by a continuous encryption function into a pair of sequences of quasi-continuous pseudorandom numbers (QCPRNs) of any desired length, and then these QCPRNs are quantized into keys. We show that CEbQ can yield a much lower key error rate than direct quantization subject to standardized randomness tests. Comparisons with other methods are also provided.
Keywords:
Quantization (signal)
Biometrics (access control)
Security
Encryption
Communication system security
Signal to noise ratio
Sensitivity
Biometric security
continuous encryption
network security
quantization
secret key generation
wireless security
Journal
IF:
9.6
Papers:
1.1W
Citations:
1.7W
