| Title |
Sequence Data-based Modulation Identification for Single and Composite Modulated Radar Signal |
| Authors |
이동은(Dongeun Lee) ; 김윤지(Yoonji Kim) ; 서동호(Dongho Seo) ; 윤우진(Woojin Yun) ; 윤동원(Dongweon Yoon) |
| DOI |
https://doi.org/10.5573/ieie.2024.61.3.3 |
| Keywords |
Radar signal modulation identification; Composite modulated radar signal; Deep learning model |
| Abstract |
Identifying the modulation scheme of collected radar signals without prior information plays an important role in electromagnetic warfare, and the information on the modulation scheme obtained through modulation identification of radar signal can be used to establish strategies and secure superiority in electromagnetic warfare. In this paper, using a deep learning model, we propose a sequence data-based modulation identification method for 37 types of single and composite modulated radar signals. In addition, we analyze the modulation identification performance according to the sequence data and the deep learning model. The proposed method generates sequence data by analyzing received radar signals in the time or frequency domain, and then using this data as input to the deep learning model, it identifies the modulation scheme. As sequence data, we consider the time domain data consisting of real and imaginary parts of the radar signal and the frequency domain data consisting of real and imaginary parts of the discrete Fourier transform. For deep learning models, we consider ResNet, DenseNet, and Inception-v3. Through computer simulations, we show that using the frequency domain data as the input of the model, the modulation identification performance is better compared to using the time domain data. It is also confirmed that the deep learning models considered in this paper show better performance than the existing deep learning model of the sequence data-based modulation identification method. Furthermore, by comparing the performance of each deep learning model, we demonstrate that the DenseNet201 exhibits the best modulation identification performance. |