中文说明：线性自适应滤波和传统非线性方法的基本概念。接着介绍核自适应滤波的数学基础-再生核希尔伯特空间理论。此处着重强调了核自适应滤波器是一个通用的函数逼近器、在自适应过程中无局部极小值而仅需要合理的计算资源。其次，本书研究了核自适应滤波系列中最简单的核最小均方算法。其三，介绍了核仿射投影算法，具体包括四个类似的算法。其四，介绍了核递归最小二乘算法和高斯过程回归理论。
English Description：
Basic concepts of linear adaptive filtering and traditional nonlinear methods. Then introduces the mathematical basis of the kernel adaptive filter - the theory of the Hilbert space of the reproducing kernel. Here, it is emphasized that the kernel adaptive filter is a universal function approximation, in which the adaptive process has no local minimum and only needs reasonable computing resources. Secondly, the most simple kernel least mean square algorithm is studied in this book. Third, the affine projection algorithm is introduced, which includes four similar algorithms. Fourth, the kernel recursive least square algorithm and Gauss process regression theory.