Indexed by:
Abstract:
针对强记忆功放的非线性问题,提出一种基于自适应扩展卡尔曼滤波与神经网络的高功放(High power amplifier,HPA)预失真算法.采用实数固定延时神经网络(Real-valued focused time-delay neural network,RVFTDNN)对间接学习结构预失真系统中的预失真器和逆估计器进行建模,扩展卡尔曼滤波(Extended Kalman filter,EKF)算法训练神经网络,从理论上指出Levenberg-Marquardt(LM)算法是EKF算法的特殊情况,并用李亚普诺夫稳定性理论分析EKF算法的稳定收敛条件,推导出测量误差矩阵的自适应迭代公式.结果表明:自适应EKF算法的训练误差和泛化误差均比LM算法更低,预失真后的邻道功率比(Adjacent channel power ratio,ACPR)比LM算法改善了2 d B.
Keyword:
Reprint 's Address:
Email:
Source :
自动化学报
ISSN: 0254-4156
Year: 2016
Issue: 1
Volume: 42
Page: 122-130
Cited Count:
SCOPUS Cited Count:
ESI Highly Cited Papers on the List: 0 Unfold All
WanFang Cited Count:
Chinese Cited Count: -1
30 Days PV: 4
Affiliated Colleges: