摘要
针对实信号去噪问题 ,系统研究了基于lk 范数正则化的去噪方法。在研究和借鉴Tikhonov正则化参数选择方法的基础上 ,给出了基于lk 范数正则化中最优正则化参数的自动选择方法 ,同时给出了正则化方法求解的迭代算法。理论分析和仿真计算结果表明 ,该方法对于加性噪声有较好的抑制能力 ,同时能够有效地保护信号中的重要特征。
For the sake of real signal de-noising, a method based on l_k norm regularization is studied. Based on parameter selection of Tikhonov regularization, automatic optimal regularization parameter selection method of l_k norm regularization is proposed. The iterative algorithm is also proposed. Theorization analysis and simulation experiment results demonstrate that this proposed method behaves well in additive noise suppression, it can efficiently preserve the important feature of the signal at the same time.
出处
《系统工程与电子技术》
EI
CSCD
北大核心
2004年第7期876-877,892,共3页
Systems Engineering and Electronics
基金
国家自然科学基金 ( 60 2 72 0 13 )
优秀博士论文作者专项基金 ( 2 0 0 14 0 )资助课题
关键词
实信号
LK范数
去噪
正则化
real signal
l_k norm
de-noising
regularization
