The use of Cann's model for modeling nonlinear circuits and systems is disc
ussed in this paper. It is shown that this model has a nonphysical behavior
for the small-signal regime in most cases (it has no some derivatives and,
correspondingly, cannot be expanded in a Taylor series) and gives incorrec
t predictions for nonlinear products. When nonlinearities are to be taken i
nto account in the simulation (intermodulation, adjacent channel power, etc
.), the use of Cann's model should be avoided or at least it should be used
with extreme care.