Blind Automatic Modulation Classification in Multipath Fading Channels

Vincent Gouldieff (1) , Jacques Palicot (1) , Steredenn Daumont (2)

1 IETR - Institut d'Électronique et des Technologies du numéRique
2 Zodiac Data Systems (Zodiac Data Systems)

Automatic Modulation Classification (AMC) has been deeply studied in the context of Cognitive Radio (CR) where it represents a required step between signal detection and demodulation. Several blind methods were proposed in the case of Additive White Gaussian Noise (AWGN) channels. In this context, Analytical M th -Power nonlinear Transformation (AMPT) was especially studied to tackle the blind AMC of digital signals. Throughout this article, we extend the study of the AMPT-based classifier to the case of imperfectly estimated multipath fading channels. To this end, the theoretical references used for the classification take the physical channel into account. Extensive simulations show the effectiveness of the proposed method compared to the existing literature.

Domains: Engineering Sciences [physics]; Signal and Image processing

Vincent Gouldieff, Jacques Palicot, Steredenn Daumont. Blind Automatic Modulation Classification in Multipath Fading Channels. IEEE Digital Signal Processing Conference (DSP), Aug 2017, London, United Kingdom. ⟨10.1109/icdsp.2017.8096116⟩. ⟨hal-01616506⟩