A wall clutter elimination method for the through-the-wall imaging radar (TWIR) with sparse array antenna is presented in this paper. Unlike the synthetic aperture radar (SAR) processing techniques, we scan the elements of the array periodically to create a range profile (RP) matrix for each transmit/receive antenna pair. Then, based on analysis of the weak fluc- tuation among RP signals, we apply independent component analysis (ICA) on each RP matrix to decompose the signal components and remove the wall clutter. The experimental results show that the proposed method can separate targets and wall clutter components, and improve the signal-to-clutter ratio (SCR) of final image effectively.