A novel modulation classification algorithm based on daubechies5 wavelet and fractional fourier transform in cognitive radio

Modulation classification is one of the most important tasks in signal waveform identification. In this paper, We address the problem of digital modulation classification in Additive White Gaussian Noise and Rayleigh fading channel. In order to classify digital modulation signal reliably, we propose an algorithm for automatic modulation recognition in this paper. We verify this algorithm using higher order statistical moments of wavelet transform and Fractional Fourier Transform as a feature set. Through this algorithm, we can discriminate 2ASK 2FSK BPSK and 16QAM without any prior signal information. In addition pre-processing and features subset selection using principal component analysis will reduce the network complexity and the recognizers performance is no significant decrease. The performance of the identification scheme is investigated through the simulations. Simulation results show that the performance of the proposed modulation classifier is superior to other existing signal classifiers.