Automatic Modulation Classification Using Involution Enabled Residual Networks

Automatic modulation classification (AMC) is of crucial importance for realizing wireless intelligence communications. Many deep learning based models especially convolution neural networks (CNNs) have been proposed AMC. However, the computation cost very high, which makes them inappropriate beyond fifth generation communication that stringent requirements on accuracy and computing time. In order to tackle those challenges, a novel involution enabled AMC scheme by using bottleneck structure residual networks. Involution utilized instead enhance discrimination capability expressiveness model incorporating self-attention mechanism. Simulation results demonstrate our achieves superior performance faster convergence speed comparing with other benchmark schemes.