TY  - JOUR
T1  - Semantic Segmentation-Based Drone Communication Signal Separation Method
AU  - Park, Do-Hyun 
AU  - Kwon, Soon-Young 
AU  - Jeong, Jinwoo 
AU  - Sim, Isaac 
AU  - Yun, Sangbom 
AU  - Seo, Junghyun 
AU  - Kim, Hyoung-Nam 
JO  - The Journal of Korean Institute of Communications and Information Sciences
PY  - 2025
DA  - 2025/1/1
DO  - 10.7840/kics.2025.50.11.1679
KW  - Electronic Warfare
KW  - Deep Learning
KW  - 
KW  - Signal Separation
AB  - Recently, the importance of accurately estimating communication signal parameters used by drones has increased for identifying drones operating for tactical purposes. However, drone communication signals often contain multiple signals transmitted concurrently within the same frequency band, making it challenging to accurately estimate the parameters of individual signals. In this paper, we propose a deep learning-based method to separate drone communication signals. The proposed approach converts the received signal into spectrogram and separates them by employing a semantic segmentation model.
Simulation results show that the proposed method accurately performs signal separation, achieving excellent performance in estimating signal parameters.