INCREASING THE STABILITY OF CRYPTOGRAPHIC ALGORITHMS IN MULTI-USER WEB RESOURCES BASED ON RANDOM NUMBER GENERATORS THAT TAKE INTO ACCOUNT THE ENTROPY OF USER BEHAVIOR

Abstract

The paper considers the problem of increasing the security of Web resources by stable crypto-algorithms based on random number generators that take into account the entropy of user behavior in a multi-user environment. The research developed an algorithm for a reliable entropy source of truly random numbers for a table generator that uses pre-prepared tables containing verified irrelevant numbers. For the software implementation of the encryption subsystem, an algorithm for the work of the tabular generator was developed, while separate modules of the random number generator were combined, namely, the algorithm for filling the buffer of the tabular generator and the algorithm for selecting a random number from it. The output data of the generator was saved in a text file, which was used as input data necessary for randomness testing. At the same time, the testing method proposed by the US National Institute of Standards and Technology - NIST based on a NIST package of statistical tests aimed at determining the degree of randomness of binary sequences generated by random number generators. The main advantage is that these tests are based on various statistical properties belonging only to random sequences. From the obtained results, it was concluded that the sequence fully satisfies the criteria of randomness. In addition, we will conduct a comparative analysis of the characteristics of the developed random number generator with classical implementations, on the basis of which its advantages are determined.