Most current deep learning (DL) based AMC methods often require a large number of labeled samples to drive itself optimization study, so as to achieve a more superior performance. However, for many AMC tasks with a small amount of data, the AMC method based on deep learning still has many challenges, which makes the deep neural network model trained on a small dataset lack effective migration effect, leading to the failure to achieve better performance effect in the testing process. In this paper, we propose a novel AMC method based on Complex-valued Signal Capsule Network (CSCN). Compared with other DL based methods, the proposed CSCN can achieve a better performance under the condition of small dataset and low SNR on the well-known deepsig dataset RML2016.10a and RML2016.10b.
Complex-Valued Signal Capsule Network for Automatic Modulation Classification
Lect. Notes Electrical Eng.
International Conference on Autonomous Unmanned Systems ; 2021 ; Changsha, China September 24, 2021 - September 26, 2021
Proceedings of 2021 International Conference on Autonomous Unmanned Systems (ICAUS 2021) ; Kapitel : 323 ; 3291-3296
18.03.2022
6 pages
Aufsatz/Kapitel (Buch)
Elektronische Ressource
Englisch
Glaucus: A Complex-Valued Radio Signal Autoencoder
IEEE | 2023
|Vehicle-Type Classification Using Capsule Neural Network
Springer Verlag | 2022
|