基于复数ICA的无线直放站反馈干扰抵消算法

Complex ICA based adaptive feedback interference cancelling algorithm for wireless repeaters

为了利用自适应滤波器抵消转发与接收天线间的反馈干扰,根据卷积型混合与独立分量分析ICA解混的原理,将反馈干扰抵消问题纳入了ICA方法的框架中。针对亚高斯复值信号的特点,依据极小化互信息的准则,提出了一种基于复数ICA的自适应算法,简要分析了算法收敛的平衡点并据此给出了新的干扰抵消器的结构。计算机仿真结果表明,相比于LMS算法,所提的算法在反馈信道辨识误差以及输出信号的误差矢量幅度方面有明显的改善。

Feedback interference cancelling by using adaptive filters is one of the core technologies of wireless repeaters. The problem was summed up into the framework of ICA in terms of the ICA demixing principle of convolutive mixtures. Aiming at complex signals＇ subgaussian feature, an adaptive algorithm based on complex ICA was proposed according to minimization of mutual information criterion. The convergence equilibrium point of the proposed algorithm was briefly analyzed and a new structure of interference cancel- ler presented accordingly. Computer simulation demonstrated that the proposed algorithm had obvious performance improvement over conventional LMS in the identification error of feedback path and of the error vector magnitude the output signals＇.