Distributed Ground-based radar target identification methods of a class nearest neighbor classifier

The present invention discloses a distributed ground radar target discrimination method based on a nearest neighbor classifier type, mainly solves the prior art problem of excessive false alarm detection. The technical scheme is: 1) a one-dimensional radar echo signals for the CFAR-norm normalization; 2). The results were taken and are aligned and selecting the training and test samples; 3) Calculation Training the average Hausdorff distance vector between the sample and determining the nearest neighbor of a class decision threshold classifier limit Thr; 4) calculating an average Hausdorff distance d between the test sample and the training samples; 5) average Hausdorff distance and the decision threshold are compared, If d≤Thr, the test sample as the goal, otherwise, the test sample is non-objective. The present invention preferably removes non-target false alarm, can be used to identify ground targets distributed.