Classification of quadrature amplitude modulated (QAM) signals via sequential probability ratio test (SPRT)

A novel approach to automatic classification of quadrature amplitude modulated (QAM) signals is presented in this research. Modulation classification has been traditionally treated as a hypothesis test problem with input signals of a fixed sample size. By formulating it as a variable sample size test problem, we propose a new classification algorithm based on the sequential probability ratio test (SPRT). It is demonstrated that the new approach has several important merits, including ease of error rate control, lower computational complexity and lower decision delay. (C) 1997 Published by Elsevier Science B.V.