Hough transform-based large micro-motion target detection and estimation in synthetic aperture radar

Target vibration or rotation produces special phase modulation of synthetic aperture radar azimuth echo signals, referred to as the micro-Doppler effect, which can provide favourable information for target detection and recognition. However, the pulse repetition frequency will be lower than the azimuth bandwidth of azimuth echo signal when the micro-motion amplitude is too large, resulting in spectral aliasing, which invalidates the conventional detection and estimation algorithm. This study proposes a new target detection and estimation algorithm for large micro-motion targets. First, the authors perform time-frequency (TF) transform on the azimuth echo signals of micro-motion targets. Then extract the TF curves in the TF distribution. After modifying the analytical expression of large micro-motion target TF curves, the target detection and parameter estimation are realised by Hough transform. The algorithm proposed in this study is applicable to the detection of both large and small micro-motion targets. It is capable of detecting multiple micro-motion targets in a single range cell simultaneously and is featured with high parameter estimation precision even if there is strong noise. Simulation results and field experimental data prove the effectiveness of the proposed algorithm and its superiority compared with the inverse radon transform algorithm.