for the quantizer, in some paper, LSB=FS/2^N, N is the total bit, FS is the full scale range,

while in some other paper (such as slgma-delta modulator), LSB=FS/(2^N-1), why?

which is correct?

and how to understand the two different difinition about LSB? in the latter, why "2[sup]N[/sup]-1" is used? if N=1, how to explain this?

please help me, thanks.