【作者】 李新苗； 李有明； 徐铁峰； 俞建定；

【Author】 LI Xin-miao,LI You-ming,XU Tie-feng,YU Jian-ding Institute of Communication Technology,Ningbo University,Ningbo,Zhejiang,315211

【机构】 宁波大学通信技术研究所；

【摘要】 针对多带正交频分复用(MB-OFDM)超宽带(UWB)系统,本文提出了一种减小信道估计运算量的快速方法。常规的时域信道估计方法利用训练序列和接收序列的时域相关来估计信道。由于涉及矩阵求逆,算法的复杂度比较高。当训练序列采用m序列时,m序列的自相关矩阵具有对角占优性。本文正是利用此特性提出两种低复杂度的超宽带信道估计新方法。仿真结果表明本文新算法复杂度低,且其性能逼近常规的时域信道估计方法。更多还原

【Abstract】 Two fast methods for reducing the complexity of channel estimation for multi-band Orthogonal Frequency Division Multiplexing (MB-OFDM) UWB are proposed in this paper.The traditional Time-domain channel estimator relies on the correlation between the received sequences and training sequences.This method requires matrix inversion,causing high computational complexity. Using diagonal dominant property of the auto-correlation matrix of training m sequences,we develop two low complexity UWB channel estimation algorithms.The correlation matrix is first decomposed as two matrices,the inversion of the matrix is then approximated from this decomposition.Comparing with traditional Time-domain channel estimator,the new algorithms have much lower computational complexity,but approximate the former estimator.Computer simulation results verify the effectiveness of the proposed methods.更多还原