Skip to Main Content
A novel recognition method is proposed to relax the heavy requirement of training set size in the radar High Resolution Range Profile (HRRP) target recognition. Firstly, the statistical characteristic of HRRP's frequency spectrum is analyzed and considered to be a wide-sense stationary process. Then a linear dynamic model is introduced to describe the frequency spectrum amplitude and an Expectation Maximization (EM) algorithm is employed to estimate model parameters. Finally, experimental results show the proposed method can obtain satisfactory recognition accuracy and rejection performance with only a few training samples.