A novel LMMSE channel estimation algorithm for multipath channel in tunnels

作者：Ding Qing Feng^*; Zheng Guo Xin; Yang Liu

来源：Journal of Convergence Information Technology, 2012, 7(21): 485-492.

DOI：10.4156/jcit.vol7.issue21.59

• Channel characteristics
• Channel estimation algorithms
• Channel impulse response
• Correlation matrix
• Information components
• Least squares
• Linear minimum mean-squared errors
• LMMSE
• LMMSE channel estimation
• Low rank approximations
• Low-rank
• Noise components
• OFDM systems
• Transform domain

A novel channel estimation algorithm of utilizing Singular Value Decomposition (SVD) and low-pass filter (LF) is proposed for OFDM systems over the tunnel multipath channel. The channel impulse response is estimated by utilizing Least Squares (LS). The correlation matrix of the channel impulse response is decomposed by SVD, so the computational complexity of the simplified linear minimum mean-squared error (LMMSE) estimator is reduced for low-rank approximation. The information component of the channel characteristics and the noise component can be separated by the fast Fourier transform (FFT). So the noise component is dramatically reduced after low-pass filter in the transform domain. Comparing the traditional LMMSE channel estimation algorithm, the SVD-LF channel estimation algorithm together with low-pass filter provides better performance between the Bit Error Rate (BER) and Mean Square Error (MSE) and requires less computational complexity.

• 出版日期2012
• 单位上海大学