.png){kind=icon}
.png){kind=icon}
Return
A Transformer and Convolution-Based Learning Framework for Automatic Modulation Classification
.png){kind=icon}
PRE
AI
DOI：10.1109/LCOMM.2024.3380623.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Automatic modulation classification (AMC) is a typical pattern classification task that is an intermediate process between signal detection and demodulation. Deep learning methods used in AMC, such as convolutional neural network (CNN) have their shortcomings. We propose a new parallel CNN transformer network (PCTNet), which not only possesses the advantages of transformer to capture long-range dependencies, but also utilizes the advantages of CNN to extract local information. PCTNet is a parallel design of CNN and transformer, with a delivery mechanism in the middle. Extensive simulation results show that our proposed PCTNet can achieve superior classification performance than traditional deep models.
Keywords:
Automatic modulation classification
signal processing
cognitive radio
deep learning
convolutional network