Deep Learning-Based Robust Automatic Modulation Classification for Cognitive Radio Networks

In this paper, a novel deep learning-based robust automatic modulation classification (AMC) method is proposed for cognitive radio networks. Generally, as network input of AMC convolutional neural networks (CNNs) images or complex signals are utilized in time domain frequency domain. terms the image that contains RGB(Red, Green, Blue) levels size may be larger than signal, which represents increase computational complexity. signal it normally used $2 \times N$ input, divided into in-phase and quadrature-phase (IQ) components. extended notation="LaTeX">$4 by copying IQ components concatenating reverse order to improve accuracy. Since amount computation complexity due size, CNN archiecture designed reduce from an average pooling layer, can enhence accuracy well. The simulation results show model higher conventional models almost signal-to-noise ratio (SNR) range.