Abstract：For Through-the-Wall Imaging Radar (TWIR), wall clutter is critical for detecting target signals behind a wall. For a system with a sparse antenna array, the lack of observation channels makes it more difficult to separate the target signals and wall clutter. On the basis of fluctuation of the range profile in real transmit/receive channels, this paper proposes to use Independent Component Analysis (ICA) on multiple down-range observations of each transmit/receive channel to remove the wall clutter. The simulation and experimental results show that the proposed method effectively separate target and clutter components, even though the signal-to-clutter ratio is only -30 dB.
