8. Conclusion

Random numbers are omnipresent in many applications. They can be found directly as a source of randomness (lotteries, games), as a means of protection (cryptography, computer system security) or as a means of accelerating certain complex calculations (randomized algorithms, digital simulation).

Two main types of random number generator currently exist: pseudo-random generators (or PRNGs), which use deterministic algorithms, and truly random generators (or TRNGs), which exploit physical noise sources with chaotic behavior. These two types of generator correspond to different needs. In the case of PRNGs, we want an efficient software implementation, and in some cases, we want to be able to "replay" this randomness (e.g. stream ciphers). In the case of TRNGs, the aim is to guarantee that the sequences produced are totally unpredictable and non-reproducible (even in...