Radar target recognition based on nonparametric density estimation

In order to solve the problem of model mismatch when using parametric approach to estimate the density of High-Resolution Range Profile (HRRP) in radar target recognition, a nonparametric method-Stochastic Learning of the Cumulative (SLC) is presented for the density estimation of HRRP. SLC uses a multiplayer network to estimate the distribution function of the training samples and obtains density by taking derivative. SLC not only describes the density function more comprehensive and accurately, but also avoids the problem of being sensitive to window width that many nonparametric approaches may suffer. Experimental results using outfield real data demonstrate the validity of the proposed learning algorithm.