The special deep neural network for stationary signal spectra classification
С. А. Субботин · 2018 14th International Conference on Advanced Trends in Radioelecrtronics, Telecommunications and Computer Engineering (TCSET) · 2018
A special deep neural network architecture for automatic classification of stationary signal spectra and the method of its training by precedents are proposed. The developed method makes possible to obtain recognition models having high generalizing properties, simplicity and interpretation, and also to reduce the time for model building. The results of experiments allow recommending the proposed method for use in practice.