Abstract
Nowadays, fifth generation (5G) network and unmanned aerial vehicle (UAV) are more and more important in the civil and military field. Only communicating correctly in 5G network and between UAVs, they can play a role in real world. Modulation classification is the premise to ensure communication in 5G network and between UAVs correctly. However, the effects of multipath fading always exists in the 5G communication environment and UAV communication channel, which leads to severe modulation classification performance and communication performance degradation. In order to resolve this problem, we proposed a novel modulation classification algorithm that can classify signals without priori information in multipath channel. The proposed algorithm makes the mean, variance, skewness and kurtosis of wavelet transform as the feature set, then uses principal component analysis (PCA) for feature subset selection, in the end neural network is used as classifier to classify signals. The simulation results show that the proposed algorithm can achieve the much better classification accuracy than the existing methods in multipath fading channels.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call