This paper presents an Automatic Modulation Identification (AMI) algorithm using Wavelet transform for distinguishing QPSK and GMSK in adaptive demodulator. The developed algorithm has been tested by introducing Additive White Gaussian Noise (AWGN) in the modulated signal. It has been found that the identification of QPSK and GMSK using histogram peaks simplifies the procedure. The performance of the proposed AMI algorithm has been computed by estimating the Bit Error Rate (BER) and reconstruction efficiency. This shows that the correct modulation identification is possible to a lower bound of 15 dB and 6 dB for GMSK and QPSK respectively. © 2007 IEEE.