Deep Learning-Based Separation Technique for Radar-Communications Overlapping Signals 


Vol. 49,  No. 5, pp. 711-717, May  2024
10.7840/kics.2024.49.5.711


PDF
  Abstract

The advancement of wireless communication technology has led to increased signal interference between signals, negatively affecting the quality and reliability of communication. To address such issues, signal processing algorithms like frequency filtering and spatial separation have been developed, yet their performance significantly degrades in environments with completely overlapping frequencies. This paper proposes a method using the deep learning model U-Net to separate overlapping signals and restore the original signals. 1D I/Q data was transformed into 2D time-frequency images through the application of the STFT algorithm, and the U-Net model was employed to segregate the overlapping signals. The goal was to restore the communication signal, and a BER of   or less was guaranteed for the BPSK signal at a SIR of -29 dB to 0 dB.

  Statistics
Cumulative Counts from November, 2022
Multiple requests among the same browser session are counted as one view. If you mouse over a chart, the values of data points will be shown.


  Related Articles
  Cite this article

[IEEE Style]

S. Lee, S. Jung, J. Jung, H. Nam, "Deep Learning-Based Separation Technique for Radar-Communications Overlapping Signals," The Journal of Korean Institute of Communications and Information Sciences, vol. 49, no. 5, pp. 711-717, 2024. DOI: 10.7840/kics.2024.49.5.711.

[ACM Style]

Si-Ho Lee, Suk-Hyun Jung, Jae-Yeon Jung, and Hae-Woon Nam. 2024. Deep Learning-Based Separation Technique for Radar-Communications Overlapping Signals. The Journal of Korean Institute of Communications and Information Sciences, 49, 5, (2024), 711-717. DOI: 10.7840/kics.2024.49.5.711.

[KICS Style]

Si-Ho Lee, Suk-Hyun Jung, Jae-Yeon Jung, Hae-Woon Nam, "Deep Learning-Based Separation Technique for Radar-Communications Overlapping Signals," The Journal of Korean Institute of Communications and Information Sciences, vol. 49, no. 5, pp. 711-717, 5. 2024. (https://doi.org/10.7840/kics.2024.49.5.711)
Vol. 49, No. 5 Index