正交FDWT与雷达数据压缩及目标识别方法

Orthogonal FDWT and Methods of Radar Data Compression and Target Identification

本文由共轭镜像滤波器构造有限维内积空间的正交多分辨分析与快速离散小波变换(FDWT),并将其用于降低雷达数据存储量,提出基于压缩数据的小波域相关与最小距离分类方法.变换域识别法计算速度快.实验结果表明,本文的变换、压缩和识别方法是有效的,小波变换域数据降维描述目标的能力强.

In this paper, making use of the orthogonal quadrature filters, we construct the orthogonal multiresolution of finite dimension inner-product space and the fast discrete wavelet transform (FDWT). and reduce radar data storage by the transform. The methods of correlation and nearest neighbour classification in wavelet domain based on the compressed data are presented. The computation is faster in the methods of transformation domain Identification. The experiment results show that the methods are efficient and the data in wavelet transformation domain can more efficiently describe targets with reduced dimension.