Multiple moving targets real-time detection in single-channel SAR using median filter

With the development of hardware, the detection of ground moving targets with single-channel synthetic aperture radar (SAR) can be achieved by a relatively higher pulse repetition frequency (PRF). This high PRF can bring adequate amount of information, but it also expands the non-clutter spectrum which affects the precision of the real-time detection by the Doppler spectrum shifting methods. In order to make sure that the single-channel SAR can achieve real-time imaging and detection of moving targets, a new method based on former Doppler shifting methods is presented in this paper. First, we calculate the azimuth spectrum by Fourier transform in segments. Then, the median filter is introduced to eliminate the high-energy, narrow-spread spikes in the real data, which makes the whole procedure to be realistic and improves the location precision of moving targets significantly. In the end, we confirm the estimated Doppler centroids for the latter imaging and detection of moving targets. The corresponding flow of the whole system is also given in this paper. The effectiveness of this method is illustrated with the simulation results based on the real data.