Neuro-fuzzy signal classifier (NFSC) for standard wireless technologies

Optimizing the coexistence performance of a cognitive radio (CR) greatly depends on the extent of its awareness of the radio environment. The more knowledge a CR can acquire about primary user (PU) systems, the better it can plan its radio activities. Ideally, it would like to classify the PU systems with respect to `known standards´ so that the documented knowledge of PU systems can be usefully utilized. Intensive research has been done in the area of signal classification but merely with respect to modulations. We present a novel neuro-fuzzy signal classifier (NFSC) to classify signals with respect to known standards. The NFSC demonstrates excellent performance by using modern A/D based wideband data acquisition as well as traditional heterodyne narrowband transceivers in real-time coexistence environments.