雷达一维高分辨距离像对噪声极为敏感。研究一种新的小波去噪算法得到高信噪比的雷达一维高分辨距离像是实现精确目标识别的关键。文中对比分析了小波模极大值去噪算法和小波阈值去噪算法,并在经典小波阈值去噪算法的基础上改进了阈值函数,提出了一种新小波阈值去噪算法。实验证明新的算法能够更好的提高雷达一维高分辨距离像的信噪比,有效抑制噪声敏感性。新算法结合了软、硬阈值算法的优点,灵活度高。

The high resolution range profile is sensitive to noise signal. The key technology of accurate target recognition is how to find a new type of de nosing algorithm of wavelet to get high SNR high resolution range profile. By measuring and analyzing de-no sing algorithm of wavelet modular maximum and denosing algorithm of wavelet threshold, the threshold function was improved and a new type of de nosing algorithm of wavelet threshold was put forward based on the classical algorithm. The experiment evi dences that the new algorithm can improve the SNR of resolution range profile better and restrain noise sensitiveness. By combining the advantage of de nosing algorithm of wavelet soft threshold and hard threshold, the new algorithm has good application value.