Weyl sequence
──────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────
top
In mathematics, a Weyl sequence is a sequence from the equidistribution theorem proven by Hermann Weyl:cite-ref-1[1]
The sequence of all multiples of an irrational α,
0, α, 2α, 3α, 4α, ...
is equidistributed modulo 1.cite-ref-2[2]
In other words, the sequence of the fractional parts of each term will be uniformly distributed in the interval [0, 1).
Contents
• See also
──────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────
In computing
In computing, an integer version of this sequence is often used to generate a discrete uniform distribution rather than a continuous one. Instead of using an irrational number, which cannot be calculated on a digital computer, the ratio of two integers is used in its place. An integer k is chosen, relatively prime to an integer modulus m. In the common case that m is a power of 2, this amounts to requiring that k is odd.
The sequence of all multiples of such an integer k,
0, k, 2k, 3k, 4k, …
is equidistributed modulo m.
That is, the sequence of the remainders of each term when divided by m will be uniformly distributed in the interval [0, m).
The term appears to originate with George Marsaglia’s paper "Xorshift RNGs".cite-ref-marsaglia-3-0[3] The following C code generates what Marsaglia calls a "Weyl sequence":
d += 362437;
In this case, the odd integer is 362437, and the results are computed modulo m = 232 because d is a 32-bit quantity. The results are equidistributed modulo 232.
See also
References
cite-note-22. ↑ citerefkuipersniederreiter2006Kuipers, L.; Niederreiter, H. (2006) [1974]. Uniform Distribution of Sequences. Dover Publications. ISBN 0-486-45019-8.
cite-note-marsaglia-33. ↑ citerefmarsaglia2003Marsaglia, George (July 2003). "Xorshift RNGs". Journal of Statistical Software. 8 (14). doi:10.18637/jss.v008.i14.