Time and Accuracy Trade-Off of LPI Radar Classification Technology Based on Time-Frequency Analysis and Deep Learning

Technology for classifying low probability of intercept (LPI) radar signals with speed and accuracy is critical cognitive communication research. We used time-frequency analysis (TFA) deep learning to classify 12 typical LPI signals. Traditional methods use the Choi-Williams distribution (CWD), which requires more than 500 times longer TFA generation time spectrogram method. In this paper, we show trade-off relationship between classification detection using a spectrogram, Wigner-Ville (WVD), CWD as training datasets. As result, model showed higher model, but prediction was 200 longer. The difference only 1 %p an SNR over −2 dB, it reached 7.5%p −10 dB. Therefore, lower shows distinct accuracy, depending on type TFA.