新的正交最小二乘方法

doi:10.3778/j.issn.1002-8331.2009.01.010

计算机工程与应用 ›› 2009, Vol. 45 ›› Issue (1): 36-38.DOI: 10.3778/j.issn.1002-8331.2009.01.010

新的正交最小二乘方法

付丽华^1,2,李宏伟²,彭熙³

1.中国地质大学地球物理与空间信息学院，武汉 430074
2.中国地质大学数学与物理学院，武汉 430074
3.华中师范大学计算机科学系，武汉 430079

收稿日期:2008-08-28 修回日期:2008-10-17 出版日期:2009-01-01 发布日期:2009-01-01
通讯作者: 付丽华

Novel orthogonal least squares algorithm

FU Li-hua^1,2,LI Hong-wei²,PENG Xi³

1.Institute of Geophysics and Geomatics，Chinese University of Geosciences，Wuhan 430074，China
2.School of Mathematics and Physics，Chinese University of Geosciences，Wuhan 430074，China
3.Department of Computer，Central China Normal University，Wuhan 430079，China

Received:2008-08-28 Revised:2008-10-17 Online:2009-01-01 Published:2009-01-01
Contact: FU Li-hua

摘要/Abstract

摘要： 正交最小二乘是一种贪婪算法，采用逐步回归建模，每一步利用搜索算法找到最小化残差的一个回归项。将其拓展为每一步搜索多个最优的回归项，从而得到一种稀疏的回归方法，并将其应用于谐波分量提取中。仿真实验说明，新方法不仅能够较为精确地逐项估计出分量的参数，而且可以对分量个数进行有效的估计。

关键词: 树结构搜索, 正交最小二乘, 谐波提取

Abstract: Orthogonal least squares（OLS） regression utilizes a greedy scheme to tune the parameters of each individual regressor term by term.To improve the performance of the greedy-scheme-based OLS algorithm，a tree structure search algorithm is constructed.At each regressor stage，this proposed OLS algorithm is realized by keeping multiple “best” regressors rather than using the“optimal”one only.The new approach is used in harmonic retrieval.Numerical results show that this new scheme is capable of estimating not only the number of the harmonic components in the given signals but also the parameters of harmonic.

Key words: tree structure search, orthogonal least squares, harmonic retrieval

付丽华^1,2,李宏伟²,彭熙³. 新的正交最小二乘方法[J]. 计算机工程与应用, 2009, 45(1): 36-38.

FU Li-hua^1,2,LI Hong-wei²,PENG Xi³. Novel orthogonal least squares algorithm[J]. Computer Engineering and Applications, 2009, 45(1): 36-38.

[1]	王刚，石守东，郑凯辉. 基于图像修复的网线绕距测量方法研究[J]. 计算机工程与应用, 2014, 50(4): 166-169.
[2]	郑明文. 改进的OLS算法选择RBFNN中心的方法[J]. 计算机工程与应用, 2009, 45(25): 52-54.
[3]	党长青,刘洁,牛分中. 基于Boosting RBF神经网络的入侵检测[J]. 计算机工程与应用, 2008, 44(15): 118-120.
[4]	苏美娟,邓伟. RBF网络的微分进化正交最小二乘算法[J]. 计算机工程与应用, 2007, 43(35): 46-48.

新的正交最小二乘方法

Novel orthogonal least squares algorithm

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 4

编辑推荐

Metrics