It's worth putting an asterisk next to this that the theory of operation isn't terribly robust. Yes, phase noise is a valid way to sample thermal noise (a true

It's worth putting an asterisk next to this that the theory of operation isn't terribly robust. Yes, phase noise is a valid way to sample thermal noise (a true random source) but it's not clear that this is what you're actually measuring on any given implementation of the TRNG. Depending on the synthesis and PnR of this module in your larger design, there's a good chance you'll have a correlation which some other part of your design especially as density goes through the roof. While it's likely to still pass statistical tests and seem chaotic, it might not be an actual TRNG without you putting in a lot of work to isolate it.