As to target recognition of the air floating corner reflector which is a new passive interference devices for the low resolution terminal guidance radar,the radar cross section (RCS) statistical features and recognition model are studied. Through theoretical derivation, the RCS model of the airfloating corner reflector and its array are constructed, and SolidWorks and electromagnetic simulation software are used to model and simulate the two types of warship targets. Considering the characteristics of RCS changing strongly with the observation angle, nine characteristic parameters, including target position, distribution and distribution characteristics are extracted by statistical means as features of followup target recognition. Based on the deep neural network, a target recognition model for the airfloating corner reflector is constructed. The experimental results show that the optimized neural network model can be used as an effective means to discriminate the air floating corner reflector which is a false target. The classification accuracy in the training set and test set is 97.4% and 94.3% respectively.