Radar HRRP Target Recognition Based on Dynamic Learning with Limited Training Data

For high-resolution range profile (HRRP)-based radar automatic target recognition (RATR), adequate training data are required to characterize a signature effectively and get good performance. However, collecting enough involving HRRP samples from each orientation is hard. To tackle the HRRP-based RATR task with limited data, novel dynamic learning strategy proposed based on single-hidden layer feedforward network (SLFN) an assistant classifier. In offline phase, used for pretraining SLFN using reduced kernel extreme machine (RKELM). online classification collected test first labeled by fusing results of current Then reliable pseudolabels as additional update parameters sequential RKELM (OS-RKELM). Moreover, improve accuracy label estimation semi-supervised method named constraint propagation-based propagation (CPLP) was developed The dynamically accumulates knowledge through learning, thereby reinforcing performance system data. Experiments conducted simulated 10 civilian vehicles real three military demonstrated effectiveness when limited.