AMC-IoT: Automatic Modulation Classification Using Efficient Convolutional Neural Networks for Low Powered IoT Devices
Muhammad Usman, Jeong–A Lee
Abstract
Automatic modulation classification (AMC) is used to identify the modulation for the received signal. IoT devices use modern communication methods which are based on multiple input multiple output (MIMO) in which the signals are received from various sources. The identification of modulation is vital. Feature based AMC methods combined with deep learning techniques has the potential to meet the latency requirement in the IoT applications. An efficient convolutional neural network based on depthwise separable convolution has been proposed to classify the modulation of the received signals. The proposed architecture has 58% less parameters than the conventional convolutional architecture and the performance is comparable.