A true random number generator based on hyperchaos and digital sound

Abstract

True random number generators (TRNG) play an important role in many fields that require unpredictable and nondeterministic random number sequences. Unlike their pseudorandom counterparts, TRNGs are more computationally expensive as they need to harvest entropy from physical phenomena. To generate high quality true random numbers at a fast rate, this paper introduces a new TRNG based on hyperchaos and digital sound. The characteristics of a hyperchaotic map such as sensitivity to initial conditions and complex behavior amplifies noise obtained by sampling environmental sound through a computer microphone. The random numbers generated are then evaluated using statistical test suites such as NIST SP 800-22, DIEHARD and ENT. Because nondeterminism cannot be proved by merely running test suites, entropy analysis is performed to determine the unpredictability of these sequences. Results show that the proposed TRNG can generate true random numbers at a high rate while maintaining strong statistical quality. In addition, the entropy source requires only an inexpensive computer microphone which is already built into many laptops and handheld devices. Therefore, the proposed generator provides a low-cost and easily obtainable option for applications that require true random numbers.