True Random Number Generator Relying on Multiple Entropy Source and Triple Oscillator for Cryptographic Purposes

The security of any cryptographic system is closely linked to the quality of the random number (RN) generator used; thus, it can be regarded as a basic component of cryptography. This paper unveils an extended ring oscillator-based true random number generator (TRNG), named triple-oscillator TRNG (TO-TRNG) which is particularly proper for smart IC cards and on-chip silicon applications. The accumulation of jitter over time, a major aspect of the ring oscillators is the source of entropy. The TO-TRNG benefits from multiplexing three oscillators, as the high-frequency oscillator, to have a highly deviated high-frequency signal. The output RN sequence is not accessible or reachable, even by configuring identical circuitry. The proposed TRNG was implemented in a 0.18 mu m TSMC CMOS process and taped out in July 2022. A random bit is available for every 3 mu s at the output. Shannon's theory is employed to evaluate and ascertain the entropy of the post-simulated random numbers. To demonstrate the higher entropy of the TO-TRNG's output RN sequence and compare it with conventional methods, simulations were conducted using the same technology.