Convolving the original sequence

Convolving theĀ A random sequence of bits. Figure Problem of assessing the quality of a random number generator Figure Problem of assessing the quality of a random number generator Despite the fact that it is impossible to absolutely accurately assess the quality of the generate random sequence there are methods for detecting correlations in it which can be use to track the incorrect operation of the RNG. Typically the following are the most wellknown batteries of statistical tests that are use to evaluate random sequences DieHard and NIST . There is a good article on Habr about how NIST tests work.

Their definition so that

Table results of passing NIST statistical tests No. Test name Determine result Result Frequency test Too many zeros or ones Successfully Checking cumulative amounts Too many zeros or ones at the beginning of the sequence Successfully Checking for holes in subsequences Deviation in the distribution of sequences of units Successfully Checking for holes A large Spain WhatsApp Number Data small number of subsequences of zeros and ones indicates that the bit stream is fluctuating too fast slow Successfully Checking matrix ranks Deviation of the distribution of matrix ranks from the corresponding distribution for a truly random sequence associate with the periodicity of the sequences.

Whatsapp Data

Checking series Uneven distribution

Successfully Spectral test Periodic properties of a sequence Successfully Checking for nonoverlapping patterns Nonperiodic patterns are too common Successfully Checking for overlapping patterns M bit sequences of ones are too common Successfully Maurers Spain Phone Number List general statistical test Compressibility regularity of a sequence Successfully Checking for random deviations Deviation from the distribution of the number of occurrences of subsequences of a certain type Successfully eleven A type of random deviation check Deviation from the distribution of the number of occurrences of subsequences of a certain type Successfully Checking approximate entropy Uneven distribution of m bit words. Low values mean high repeatability Successfully.

Leave a Comment