A Low Complexity Channel Estimation Algorithm for TDS-OFDM System

LMMSE algorithm is one of the best linear channel estimations for time domain synchronous orthogonal frequency division multiplexing (TDS-OFDM) system. The research on the computational complexity reduction of LMMSE algorithm remains to be a challenging topic. Since the combination of fast Fourier transform with singular value decomposition (FFT-SVD) simplify the channel autocorrelation matrix computation in frequency domain, an improved LMMSE algorithm based on FFT-SVD is proposed in this paper to achieve the similar MSE performance with lower computational complexity. The simulation results show that by using the improved LMMSE algorithm, the time spending in calculation can be reduced to 70% compared with the conventional one when adopting 256 PN sequence in TDS-OFDM system over Rayleigh fading channels.