Automatic Classification of Modulations in the Flat Fading HF Channel Based on Watterson Model

The problem of automatic modulation classification in the flat fading HF channel based on Watterson Model is considered in this paper. A novel classification algorithm, based on Feature-vector composed of 9 features of 4 categories, such as, the spectral lines of quadratic and quartic nonlinearities, is proposed. The probabilities of correct classification are provided which can achieve more than 90% at SNRs as low as 7 dB.