计算机工程与应用 ›› 2007, Vol. 43 ›› Issue (16): 210-212.
• 工程与应用 • 上一篇 下一篇
陈凯亚，王敏锡
收稿日期:
修回日期:
出版日期:
发布日期:
通讯作者:
CHEN Kai-ya，WANG Min-xi
Received:
Revised:
Online:
Published:
Contact:
摘要：
提出了一种新的针对记忆非线性功率放大器的支持向量机（SVM）预失真器。通过对其建模中采用径向基核函数和多项式核函数所表现出的性能特点进行分析，为核函数的选取提供了参考。采用以多项式为核函数的SVM对3种典型的记忆非线性功率放大器模型进行线性化仿真，结果表明了该方法的有效性和鲁棒性。
Abstract: A novel predistorter model based on Support Vector Machine（SVM） is proposed to linearlize nonlinear Power Amplifier（PA） with memory.Kernel function is an important part of the SVM，different choice about kernel can lead to different result.When acting as a part of SVM predistorter，the polynomial kernel performs better than the Gaussian kernel does.Simulation on linearlizing three kinds of PA models confirms that the SVM predistorter which adopts polynomial kernel is effectiveness and robustness.
陈凯亚，王敏锡. 基于支持向量机的记忆功率放大器预失真[J]. 计算机工程与应用, 2007, 43(16): 210-212.
CHEN Kai-ya，WANG Min-xi. Predistortion method for PA with memory based on Support Vector Machine[J]. Computer Engineering and Applications, 2007, 43(16): 210-212.
0 / 推荐
导出引用管理器 EndNote|Ris|BibTeX
链接本文: http://cea.ceaj.org/CN/
http://cea.ceaj.org/CN/Y2007/V43/I16/210
