This article presents our initial results in deep learning for modulation recognition with the signal IQ-eyes diagrams. we design a representative convolution neural network to distinguish between eight modulation classes which shows more promising when compared to more traditional likelihood-based or feature-based techniques. Furthermore, we discuss the impact of frequency offset and influence of network parameters on recognition performance though the recognition accuracy. Moreover, the method based on multi-terminal processing can mine the intrinsic characteristics of the signal more completely, which reduces the requirement of network and makes the network more robust.
Paper
Full text
Modulation Recognition Based on IQ-eyes Diagrams and Deep Learning
Semantic Scholar · Computer Science · 2019
Abstract
This article presents our initial results in deep learning for modulation recognition with the signal IQ-eyes diagrams. we design a representative convolution neural network to distinguish between eight modulation classes which shows more promising when compared to more traditional likelihood-based or feature-based techniques. Furthermore, we discuss the impact of frequency offset and influence of network parameters on recognition performance though the recognition accuracy. Moreover, the method based on multi-terminal processing can mine the intrinsic characteristics of the signal more completely, which reduces the requirement of network and makes the network more robust.