RIA: Innovative Design and Analysis of Blind Adaptive Equalization Systems

在高速数字通信系统中，盲自适应均衡器是消除非理想和带宽受限信道引起的线性失真的重要部件。通过在不借助训练序列的情况下消除线性信道失真，盲均衡使得发射机不需要为了新调谐的接收机而中断正常的数据传输来产生训练信号。尽管现有的盲均衡方案过多，但很少有方案表现出期望的全局收敛特性。许多现有的盲均衡器容易出现不期望的局部收敛，对应于很少或没有消除码间干扰，如主要研究者和其他人先前的工作所示。此外，现有的许多算法的收敛速度都很慢。本研究的一个主要目的是通过设计新的和更好的盲均衡算法来提高盲均衡系统的性能。在开发、分析和测试快速且全局收敛的均衡算法时，将使用各种线性和非线性约束。将利用PAM/QAM调制信道输出的循环平稳特性来产生盲均衡器的创新设计。此外，各种随机动态系统分析工具(如概率和随机过程、谱分析、平均和稳定性理论)将继续在本研究中使用，就像在以前的许多自适应系统研究中一样。然而，与以往的大多数研究不同的是，基于信道输出的循环平稳特性，循环平稳信号分析也将被用于盲均衡器的设计中。这项研究旨在扩大盲自适应均衡概念的实用性和广播数字通信系统的应用，并有望对自适应系统在自适应系统控制、机器人、阵列处理、机械和航天工程以及地球物理信号处理等领域的应用产生重大的直接和间接影响。

Blind adaptive equalizers are important elements in high data rate digital communication systems to remove linear distortions incurred by nonideal and bandlimited channels. By removing linear channel distortions without the aid of a training sequence, blind equalization makes it unnecessary for the transmitter to interrupt the normal data transmission in order to generate a training signal for the sake of a newly tuned receiver. Despite the plethora of existing blind equalization schemes, few exhibit the desired global convergence property. Many existing blind equalizers are prone to undesirable local convergence corresponding to little or no intersymbol interference removal, as shown in the previous works by the principal investigator and others. In addition, the convergence rates of many of the existing algorithms are very slow. A principal objective of this research is to improve the performance of blind equalization systems by designing new and better blind equalization algorithms. Various linear and nonlinear constraints will be utilized in developing, analyzing and testing fast and globally convergent equalization algorithms. The cyclostationary nature of the PAM/QAM modulated channel output will be exploited to generate innovative designs of blind equalizers. In addition, various stochastic dynamic systems analysis tools (such as probability and stochastic processes, spectral analysis, averaging and stability theory) will continue to be exploited in this study, as it has been in many previous adaptive system studies. However, unlike most of the previous studies, cyclostationary signal analysis will also be utilized in the design of blind equalizers based on the cyclostationarity of the channel output. The proposed research is aimed at expanding the utility of the concept of blind adaptive equalization and applications of broadcast digital communication systems, and is expected to have significant direct and indirect impact on the application of adaptive systems in areas such as adaptive system control, robotics, array processing, mechanical and aerospace engineering, and geophysical signal processing.