CLASSIFICATION OF VOICEBAND DATA SIGNALS USING THE CONSTELLATION MAGNITUDE

In this paper, we discuss the real time classification of high speed ( greater than or equal to 4800 b/s) voiceband data signals using blind or self-recovering equalization techniques. After equalization, the si gnals are classified, by-means of their constellation magnitude, using a new classification scheme which computes the probability that the u nknown test signal belongs to each of several predetermined classes. H euristic rules are applied to these probability functions to perform t he actual signal detection. Even in the presence of added channel impa irments, the use of this new approach resulted in an overall classific ation accuracy of greater than 98%, measured using 5100 test segments collected from 6 classes of signals produced by V.29, V.32, and V.33 m odems. Classification times range from median values of 0.090 s and 0. 219 s for the V.29 modem at 4800 and 7200 b/s, respectively, up to a w orse case time of 3.33 s for the 9600 b/s default class, chosen only i f no other signal is detected within a prescribed amount of time.