Software random number generation method with dispersion mechanism

Abstract

The invention provides a software random number generation method with a dispersion mechanism. The method includes the steps of firstly, obtaining an initial random number seed which is a read system time value and accounts for N bytes; secondly, obtaining a process random number for the random number seed through a linear congruence algorithm, and taking the result of the low N bytes and abandoning the overflow bit if overflow occurs in the process of the linear congruence algorithm; thirdly, conducting dispersion on the generated process random number, namely, dividing every N bytes of the process random number into a group, inquiring the value of each byte in an enquiry table, conducting replacement after each byte is inquired, and sequentially combining the results obtained through replacement to obtain a result random number; fourthly, outputting the result random number. Through the software random number generation method, the data dispersion mechanism is added to the loop iteration process, the randomness of software random data is enhanced, the possibility that the software random number generation method is cracked is greatly lowered, and therefore safety of products is improved.

Description

technical field [0001] The invention relates to the technical field of computer communication, in particular to a software random number generation method with a decentralized mechanism. Background technique [0002] Random numbers play a very important role in the field of security algorithms. According to their implementation methods, they can be divided into two categories: hardware random numbers and software random numbers. Generally, the performance of hardware random numbers is better than that of software random numbers. The performance of so-called random numbers can be tested by using a large number of random number samples in the NIST (National Institute of Standards and Technology) random number test tool (a currently more authoritative random number test tool). test tool) to conduct the test, and evaluate the pass rate of the test result of the tool. However, in the prior art, the production device of the hardware random number requires relatively high cost, wh...

AI Technical Summary

Problems solved by technology

[0003] Since the common algorithm of software random numbers is the linear congruence algorithm, the disadvantage of this algorithm is that it is easy to generate regularity in the process of loop iteration, that is, the randomness is weakened.

Images