首页> 外文期刊>High Technology Letters >Blind Equalization Using a Hybrid Algorithm of Multilayer Neural Network
【24h】

Blind Equalization Using a Hybrid Algorithm of Multilayer Neural Network

机译:多层神经网络混合算法的盲均衡

获取原文
获取原文并翻译 | 示例
           

摘要

A novel blind equalization scheme based on multilayer neural network and Higher Order Cumulants (HOC) is proposed in the paper. The training of the neural network uses a new hybrid algorithm which has strict convex character (after a threshold) and converges much faster than the CMA algorithm. The inverse channel is built on the basis of the estimated channel and the training of neural network. The scheme can be used in nonlinear and time varying channel and to deal with PAM or QAM signals. Simulation results show that it performs well for blind equalization.
机译:提出了一种基于多层神经网络和高阶累积量(HOC)的盲均衡方案。神经网络的训练使用一种新的混合算法,该算法具有严格的凸特征(在阈值之后),并且收敛速度比CMA算法快得多。逆通道是在估计通道和神经网络训练的基础上构建的。该方案可用于非线性和时变信道,并处理PAM或QAM信号。仿真结果表明,该算法在盲均衡中表现良好。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号