A novel iris and chaos-based random number generator

A novel kind of iris and chaotic-based random number generator (ICRNG) is developed from the unique randomness and unpredictability of iris. This method combines biometric feature extraction and random number generation in a novel way. Firstly, we get the non-deterministic source-iris image by iris acquisition equipment, then, we use chaotic function to eliminate the similar pattern in iris from the same person and get the unpredictable random sequence. We use auto-correlation function method and correlation coefficients method to show that ICRNG is linearly independent, then, we use BDS statistical test method to verify ICRNG is nonlinearly independent. Finally, the randomness of ICRNG is verified by histogram analysis, information analysis, sensitivity analysis, FIPS 140-2 tests and NIST SP 800-22 tests. Hence, ICRNG possess satisfactory performance and can be implemented on common PC platform.

► We extract the randomness of iris to develop a true random number generator.
► We show ICRNG is linearly and nonlinearly independent, so it is unpredictable.
► ICRNG can pass the famous FIPS 140-2 tests and NIST SP 800-22 test.
► The random sequence generated by ICRNG is absolutely different, and it cannot be reproduced.