在雷达目标识别中，由于雷达目标一维像特征提取容易，而且具有较好的识别效果，因此近来逐渐得到应用．但由于它是以散射点分布模型为依据的，对姿态变化非常敏感，为了提高对姿态变化范围较大的目标一维距离像的识别能力，在本文中，我们基于子波变换的多分辨分析理论及树形分层理论，提出一种级联SOM网络结构用于高分辨一维距离像的识别，并获得了令人满意的结果． In the radar target recognition, one-dimensional image feature extraction of radar target is easy and has good recognition effect and has been gradually applied recently. However, since it is based on the scattering point distribution model and is very sensitive to the attitude change, in order to improve the recognition ability of the target one-dimensional distance image with a large range of attitude change, in this paper, we analyze the multi-resolution based on wavelet transform Theory and tree layering theory, a cascade SOM network structure is proposed for high-resolution one-dimensional distance image recognition and satisfactory results are obtained.