Radar HRRP Target Recognition Based on Variational Auto-encoder with Learnable Prior

In this paper, we purpose a variational auto-encoder with learnable prior to learn the latent features of high-resolution range profile (HRRP) samples for downstream task, target recognition in this paper. The measured data based experiments illustrate that compared with the traditional and mostly recent HRRP recognition methods, the proposed one is not only noise-robust and data insensitive but also achieves state-of-the-art performance.