Best Paper Award
A Process and Temperature Tolerant Oscillator-Based True Random Number Generator
Takehiko AMAKI，Masanori HASHIMOTO，Takao ONOYE
[Trans. Fundamentals., Dec. 2014]

Takehiko AMAKI
Masanori HASHIMOTO
Takao ONOYE		  
  High-quality random number generation is essential for security. True random numbers are produced from a physical random source. All bits in bit streams are independent of the other bits and the probabilities of 1/0 occurrences are identical. Because true random numbers cannot be predicted by computational methods, they are very advantageous for security purposes. For example, they are used as the keys and initial vectors for the cipher block chaining (CBC) mode in common key cryptosystems. Challenge-and-response authentication also requires true random numbers for validation.

  This work focuses on oscillator-based true random number generators (TRNG) that produce physical random numbers from the jitter of oscillators. Generally, the 0/1 probability of TRNG output varies according to manufacturing variability and temperature fluctuations. To overcome this issue, this paper presents a TRNG that dynamically unbiases the 0/1 probability. The proposed TRNG automatically adjusts the duty cycle of a fast oscillator to 50%, and generates unbiased random numbers while tolerating process variations and dynamic temperature fluctuations. A prototype chip of the proposed TRNG was fabricated with a 65nm CMOS process. Measurement results show that the developed duty cycle monitor obtained a probability of ‘1’ 4,100 times faster than the conventional output bit observation, or estimated the probability with 70-fold higher accuracy. The proposed TRNG adjusted the probability of ‘1’ to within 50±0.07% in five chips within a temperature range of 0°C to 75°C. Consequently, the proposed TRNG passed the NIST and DIEHARD tests at 7.5Mbps with 6,670μm2 area. The proposed TRNG can contribute to enhancing the reliability of security systems such as authentication systems and cryptosystems.

Close