How to do random testing

Pseudorandom Number Generators (PRNGs) are critical tools for implementing secure systems. It is a natural question: how do we know PRNGs are good? One proposal is to use a suite of statistical tests to evaluate the "goodness" of PRNGs. Our project looks at the "goodness" of the suites themselves. We report on investigating the complications of running randomized testing when the "goodness" of that randomization is essentially under investigation.