• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文学位 >Performance Analysis and Design of Post-demodulator Adaptive Filters for FSK Signals in a Multipath Environment.
【24h】

Performance Analysis and Design of Post-demodulator Adaptive Filters for FSK Signals in a Multipath Environment.

机译：多径环境中FSK信号的解调后自适应滤波器的性能分析和设计。

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

Frequency-Shift Keying (FSK) continues to be an important component of modern communication applications, and development of lower impact methods for mitigating performance degradation due to multipath propagation are needed in many application areas. Equalization of multipath channels for FSK signals is typically accomplished with an adaptive filter at the receiver before demodulation such as constant modulus algorithm adaptive (CMA) filter. Although the use of CMA filters to equalize multipath channels for FSK signals is well understood, empirical tests have suggested that in some environments a Least-Mean-Square (LMS) filter positioned after the demodulator could provide similar performance improvement. Motivated by a desire to reduce computation, a post-demodulator (post-D) adaptive filter positioned after the demodulator has been proposed, and simulations using multipath models derived from physical channel measurements show promise. However due to the demodulation nonlinearity between the channel and the filter, conventional linear analysis techniques can not be used, and prior to this work there was no analytical or systematical analysis of this problem.;This thesis characterizes the performance of post-D filters for Continuous Phase Modulation (CPM) signals, a general form of FSK signals, received in multipath environments. In addition to addressing the nonlinear interaction of system parameters, it also addresses the issue that the often assumed alignment of Bit-Error-Rate (BER) with Mean-Square-Error (MSE) performance is not valid and that minimizing the MSE does not reliably minimize the BER. Theoretical analysis showing the relationship of system parameters to the nonlinear system is presented which explains a number of nonintuitive observations of the non-Gaussian statistics of the demodulator output in multipath channels and why a converged LMS filter produces a BER higher than the minimum achievable by a linear filter.;According to the analysis, two modifications of the post-D filter are developed. One, the selected update LMS (SULMS) filter applies a sigmoid function to the error used for adaptation and the other limits the range of values on the input to the adaptive filter. Both of the modified filters are shown to have improved BER performance compared to an LMS algorithm, and the SULMS reliably improves BER performance. The analysis of all simulations provides a simplified interpretation of the interaction of system and channel model parameters which may form the basis for more general use of a post-D adaptive filter.;There are three contributions of this work. The nonlinear system with the adaptive post-D filter is simulated under a much wider variety of parameters than previously reported. This reveals previously undocumented relationships of some parameters, and shows that minimizing the MSE or cluster variance in general does not minimize BER. In some cases it increases BER. The second contribution is the analysis of the histogram of the demodulator output and the impact of the adaptive filter on the histogram which leads to an understanding of why the minimum MSE does not provide a minimum BER. The third contribution is the development of a modified adaptive algorithm that improves BER without a significant increase in adaptation time or computation. This method can be used in decision directed mode as well as training mode.
机译：频移键控（FSK）仍然是现代通信应用程序的重要组成部分，在许多应用领域中，需要开发较低影响的方法来减轻由于多径传播而导致的性能下降。 FSK信号的多径信道均衡通常是在解调之前使用接收机处的自适应滤波器（例如恒模算法自适应（CMA）滤波器）来实现的。尽管使用CMA滤波器均衡FSK信号的多径信道已广为人知，但经验测试表明，在某些环境中，位于解调器之后的最小均方（LMS）滤波器可以提供类似的性能改进。出于减少计算的需求，提出了在解调器之后放置一个后解调器（后D）自适应滤波器，并且使用从物理信道测量中得出的多径模型进行仿真显示了希望。但是由于信道和滤波器之间的解调非线性，不能使用传统的线性分析技术，在进行这项工作之前，还没有对该问题进行分析或系统的分析。在多径环境中接收到的连续相位调制（CPM）信号，这是FSK信号的一种通用形式。除了解决系统参数的非线性相互作用之外，它还解决了以下问题：通常假定的误码率（BER）与均方误差（MSE）性能的对齐是无效的，并且最小化MSE无效可靠地降低BER。进行了理论分析，显示了系统参数与非线性系统之间的关系，该理论分析解释了多径通道中解调器输出的非高斯统计量的许多非直观观察，以及为什么收敛的LMS滤波器产生的BER高于通过AFM可获得的最小值。根据分析，对后置D滤波器进行了两种修改。一种是，选定的更新LMS（SULMS）过滤器将S型函数应用于用于自适应的误差，另一种是限制输入到自适应过滤器的值的范围。与LMS算法相比，这两个改进的滤波器均显示出改善的BER性能，并且SULMS可靠地提高了BER性能。对所有仿真的分析为系统和信道模型参数之间的相互作用提供了简化的解释，这可以为更广泛地使用D后自适应滤波器奠定基础。这项工作有三点贡献。具有自适应后置D滤波器的非线性系统是在比以前报道的参数更广泛的范围内模拟的。这揭示了一些参数以前未记录的关系，并且表明最小化MSE或群集方差通常不会最小化BER。在某些情况下，它会提高BER。第二个贡献是对解调器输出的直方图的分析以及自适应滤波器对直方图的影响，这有助于理解为什么最小MSE不能提供最小BER。第三个贡献是改进了自适应算法的开发，该算法改进了BER，而又不显着增加自适应时间或计算量。此方法可用于决策指导模式和训练模式。

著录项

  • 作者

    Lee, Shu-Ting.;

  • 作者单位

    Santa Clara University.;

  • 授予单位 Santa Clara University.;
  • 学科 Engineering Electronics and Electrical.
  • 学位 Ph.D.
  • 年度 2011
  • 页码 123 p.
  • 总页数 123
  • 原文格式 PDF
  • 正文语种 eng
  • 中图分类
  • 关键词

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