D-GF-CNN Algorithm for Modulation Recognition

Abstract

\n This paper presents a novel modulation recognition algorithm based D-GF convolutional neural networks, named as D-GF-CNN algorithm. Firstly, a asynchronous delay sampling (ADS) technique is introduced. Via the defined ADS, the received signal is converted into asynchronous delay histogram (ADH). The ADH of different modulation signal has distinct characteristics, which provides great convenience for the neural network to identify the modulation mode. Then, the pixel point matrix of histogram is convolved with the dilated convolution kernel of the convolutional neural network, and the automatic extraction of signal features is completed so that the manual feature extraction processing can be effectively avoid. According to the optimization theory, a novel GF regularization function is given, which can improve the constraint ability of the loss function on the weight and effectively weaken the influence of network over-fitting on the modulation recognition accuracy. Theoretical analysis and simulation experiments show that the proposed algorithm can offer several advantages, such as automatically extract features, effectively prevent network over-fitting and improve recognition accuracy, etc.