This paper proposes a radar aerial target recognition method based on the Mamba neural network architecture. The design includes a deep network structure with 5 Mamba blocks, fully utilizing the advantages of its state space model and temporal mixing layers in sequence modeling, proposed model effectively captures the temporal features and target structural information contained in high-resolution one-dimensional range profiles(HRRP) of three typical aerial targets, achieving high-precision target recognition. Experimental results show that compared to other deep learning methods, the proposed method demonstrates significant improvements in recognition efficiency, exhibiting good practical value.