Abstract: It is necessary to research nonlinear channel equalization to reduce Inter-Symbol-Interference（ISI） in the high-
rated wireless communication area. Adaptive equalization algorithm of nonlinear channel is studied in the Reproducing Kernel Hilbert Space（RKHS）. The structure of equalizer is proposed on nonlinear Wiener model and Kernel Affine Projection Algorithms（KAPA） is derived based on kernel method in RKHS and Affine Projection Algorithms（APA）. A simulation for the proposed adaptive algorithm is made by Monte Carlo method and comparisons are drawn between the algorithm and the others in the paper on convergence performance, bit error rate, traceability, computational complexity and so on. Simulation results show that the proposed algorithm can greatly reduce Bits Error Ratio（BER） without increasing computation complexity and is well suited for time-varying, nonlinear channel equalization.

Key words: nonlinear channel equalization, reproducing kernel Hilbert space, kernel method, Winner model, Affine Projection Algorithm（APA）, Kernel Affine Projection Algorithm（KAPA）, Monte Carlo method

摘要： 在高速无线通信领域，为消除码间干扰（ISI）必须研究非线性信道均衡技术。基于再生核希尔伯特空间（RKHS）研究非线性信道的自适应均衡算法。首先基于非线性维纳模型提出均衡器的结构，基于RKHS引入核方法，与仿射投影算法（APA）相结合推导出核仿射投影算法（KAPA），再通过引入松弛因子得到改进的KAPA算法。用蒙特卡罗法对提出的自适应算法进行仿真，从收敛性能、误码率（BER）、跟踪能力、计算复杂度等方面与其他算法做比较。在不增加计算复杂度的情况下，极大降低了误码率，非常适合时变非线性信道均衡的应用。

关键词: 非线性信道均衡, 再生核希尔伯特空间, 核方法, 维纳模型, 仿射投影算法, 核仿射投影算法, 蒙特卡罗方法