Computer Engineering and Applications ›› 2012, Vol. 48 ›› Issue (6): 106-109.

• 数据库、信号与信息处理 • Previous Articles     Next Articles

Research on channel estimation of MIMO system based on FastICA

NIU Dezhi, CHEN Changxing, WANG Shuzhao, WANG Yinhe, ZHANG Mingliang   

  1. NIU Dezhi, CHEN Changxing, WANG Shuzhao, WANG Yinhe, ZHANG Mingliang
  • Received:1900-01-01 Revised:1900-01-01 Online:2012-02-21 Published:2012-02-21

FastICA在MIMO信道估计中的应用研究

牛德智,陈长兴,王曙钊,王寅贺,张明亮   

  1. 空军工程大学 理学院 西安 710051

Abstract: Channel Estimation(CE) problem of MIMO wireless communication system is studied. By analyzing for the consistency between CE and BSS(Blind Source Separation), it proposes to apply FastICA—typical algorithm of BSS to CE problem for MIMO system, and signal presentation form is expanded from real number to complex. Aiming at the problem of channel parameters’ non-correspondence caused by order uncertainty of BSS, it proposes to modify the solved mixed matrix using correlation coefficient to relocate place, further more, corresponding solution of CE is given by method of computing inverse matrix. The simulation results show the effectiveness of means.

Key words: Multiple Input Multiple Output(MIMO), channel estimation, FastICA, correlation coefficient

摘要: 研究了MIMO无线通信系统的信道估计问题。通过对信道估计和盲分离问题的一致性分析,提出将盲分离的典型算法FastICA应用于MIMO信道估计问题中,并将信号表征形式从实数域拓展到复数域。针对盲分离次序的不确定性以至造成的估计信道参数不对应的问题,提出了相关系数重新定位法对所得的解混矩阵进行修正,最终用矩阵求逆的方法求得信道估计问题的对应解。仿真实验结果证明了方法的有效性。

关键词: 多输入多输出(MIMO), 信道估计, FastICA算法, 相关系数