Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Real-time-safe random number generator. #130

Merged
merged 1 commit into from
Aug 27, 2024
Merged

Real-time-safe random number generator. #130

merged 1 commit into from
Aug 27, 2024

Conversation

shuhaowu
Copy link
Contributor

@shuhaowu shuhaowu commented Aug 27, 2024

Xorshift random number generator is a constant-time pseudorandom
generator that's real-time-safe.

We also create an "uniform" distribution function that converts the
random number generator (which generates a random 64 bit sequence) to a
number in [0, 1). This uniform distribution is guaranteed to be constant
time but is not 100% uniform as it does not regenerate a number if the
underlying generator generates its max value. In the case where the
underlying generator generates its max value, this distribution function
will generate 0.0. This means the chance of generating 0.0 is slightly
higher than every other value, making the distribution not truly
uniform. The non-uniformity should be very small with respect to the
Xorshift64 generator so it should work for non-cryptographic purposes
(which you probably shouldn't be doing in real-time anyway).

Xorshift random number generator is a constant-time pseudorandom
generator that's real-time-safe.

We also create an "uniform" distribution function that converts the
random number generator (which generates a random 64 bit sequence) to a
number in [0, 1). This uniform distribution is guaranteed to be constant
time but is not 100% uniform as it does not regenerate a number if the
underlying generator generates its max value. In the case where the
underlying generator generates its max value, this distribution function
will generate 0.0. This means the chance of generating 0.0 is slightly
higher than every other value, making the distribution not truly
uniform. The non-uniformity should be very small with respect to the
Xorshift64 generator so it should work for non-cryptographic purposes
(which you probably shouldn't be doing in real-time anyway).
@shuhaowu shuhaowu merged commit a23ffef into master Aug 27, 2024
3 checks passed
@shuhaowu shuhaowu deleted the xorshift branch August 27, 2024 02:34
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

1 participant