In this paper, we present a fuzzy logic modulation classifier that works in
nonideal environments in which it is difficult or impossible to use precis
e probabilistic methods. We first transform a general pattern classificatio
n problem into one of function approximation, so that fuzzy logic systems (
FLS's) can be used to construct a classifier; then, me introduce the concep
ts of fuzzy modulation type and fuzzy decision and develop a nonsingleton f
uzzy logic classifier (NSFLC) by using an additive FLS as a core building b
lock. Our NSFLC uses two-dimensional (2-D) fuzzy sets, whose membership fun
ctions are isotropic so that they are well suited for a modulation classifi
er (MC). We establish that our NSFLC, although completely based on heuristi
cs, reduces to the maximum-likelihood modulation classifier (ML MC) in idea
l conditions. In our application of NSFLC to MC in a mixture of alpha-stabl
e and Gaussian noises, we demonstrate that our NSFLC performs consistently
better than the ML MC and it gives the same performance as the ML MC when n
o impulsive noise is present.