...
  • 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>Journal of the Institution of Engineers (India) >Generalized Neuron Based Non-linear Channel Equalization and Impact of Various Error Functions on the Training Performance of the Equalizer
【24h】

Generalized Neuron Based Non-linear Channel Equalization and Impact of Various Error Functions on the Training Performance of the Equalizer

机译：基于广义神经元的非线性通道均衡以及各种误差函数对均衡器训练性能的影响

获取原文
获取原文并翻译 | 示例
           
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Equalization is necessary in digital communication systems to mitigate the effects of inter symbol interference (ISI) and various other noise sources. In this paper, a reduced complexity digital communication channel equalizer is implemented using a single generalized neuron (GN). Simulation results show that this reduced complexity simple architecture equalizer gives performance approaches to the Bayesian optimal bit error rate (BER) performance. Training performance of GN based non-linear channel equalizer has been further studied with various error functions and shown that training performance may further be improved by selecting suitable error functions. Cauchy's error function can also be a good choice for non-linear channel equalization.
机译：在数字通信系统中，必须进行均衡以减轻符号间干扰（ISI）和各种其他噪声源的影响。在本文中，使用单个广义神经元（GN）实现了降低复杂度的数字通信信道均衡器。仿真结果表明，这种降低复杂度的简单体系结构均衡器为贝叶斯最佳误码率（BER）性能提供了性能解决方案。对具有各种误差函数的基于GN的非线性信道均衡器的训练性能进行了进一步研究，结果表明，通过选择合适的误差函数可以进一步提高训练性能。 Cauchy的误差函数对于非线性通道均衡也可能是一个不错的选择。

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