I am trying to understand the actual implications for the following statement regarding a 16-bit random number generator for the tag from the EPC Generation 2, UHF RFID specification, section 6.3.2.7. that states:
Probability of a single RN16: The probability that any RN16 drawn from the RNG has value RN16 = j, for any j, shall be bounded by $$0.8/2^{16} < P(RN16 = j) < 1.25/2^{16}$$
I realized this morning that I actually have no idea what that statement really means, but it seems one can have less than random, random number generator.
I made a 16-bit random number generator on silicon that passed statistical tests; however, looking back at the specification, it seems that I probably could have saved some power and cut some corners, but I honestly don't know how I would have turned those parameters into an engineering decision.
This gets me to the question: How does the above equation affect an allowable decrease in randomness from a truly random number generator?
edit: That statement is just so weird. If I counted up from 0 to 65535 and wrapped around to 0 again, I'd satisfy that condition, and that's not random at all.