Random number generators (RNGs) play a crucial role in cryptographic schemes. If the generated random numbers exhibit patterns or are predictable, it can lead to vulnerabilities and compromise the security of cryptogr...
详细信息
Random number generators (RNGs) play a crucial role in cryptographic schemes. If the generated random numbers exhibit patterns or are predictable, it can lead to vulnerabilities and compromise the security of cryptographic protocols, including confidentiality, integrity, and authenticity. However, not all RNGs are suitable for cryptographic applications. Pseudo-random number generators (PRNGs), which are based on mathematical formulas, are highly vulnerable to attacks and can be predictable. Cryptographically Secure Pseudo-random Number Generators (CSPRNGs) offer improved security but require more resources and can still be predictable if the seed is known. True Random Number Generators (TRNGs) extract randomness from physical sources, such as atmospheric noise, thermal noise, or radioactive decay, making them truly unpredictable. memory Physically Unclonable Functions (PUFs) are promising candidates for TRNGs as they leverage the random silicon fabrication process to generate inherent randomness. The objective of this work is to improve an MRAM-based TRNG by manipulating the analog responses. We propose both hardware and software implementations of TRNG schemes. To evaluate the randomness and quality of the generated sequences, we subject them to the statistical test suite from the National Institute of Standards and Technology (NIST) designed for assessing random and pseudo-random numbers. Additionally, we introduce a post-processing method that involves XORing the generated numbers with pseudo-random numbers to enhance the randomness further and strengthen the overall security of the TRNG.
暂无评论