| Title |
Automatic Modulation Recognition based on Cyclic Moment using Convolution Neural Network |
| Authors |
최영익(Young-Ik Choi) ; 김상수(Sang-Su Kim) ; 오승섭(Seung-Sup Oh) ; 고재헌(Jae-Heon Ko) ; 장연수(Yeon-Soo Jang) |
| DOI |
https://doi.org/10.5573/ieie.2021.58.11.79 |
| Keywords |
Automatic recognition modulation; Cyclic moment; Convolution neural network |
| Abstract |
In this paper, we have presented a method for automatically recognizing and classifying modulation in wireless communication systems using a deep learning model. We extracted Cyclic Moment feature values using the cyclostationarity of signals that have robust characteristics in the channel environment such as noise, timing offset, frequency, and phase offset, and applied a CNN(Convolution Neural Network) model-based classifier. Experiments were performed under SNR(Signal-to-Noise Ratio) and Rayleigh fading channel conditions to confirm the performance of the proposed model. It showed an average recognition performance of 80.81% under the minimum SNR 0 dB condition, and confirmed the possibility of applying an automatic modulation recognition algorithm with a CNN structure using that feature value. |