Performance of decision-directed channel estimation using effective path selection in time domain for uplink DFT-precoded OFDMA

This paper proposes decision-directed channel estimation (DDCE) applying threshold based effective path selection in the time domain for a time-multiplexed reference signal (RS) in discrete Fourier transform (DFT)-precoded orthogonal frequency division multiple access (OFDMA) using a frequency domain equalizer (FDE). In the proposed method, the threshold based effective path selection in the time domain generates an accurate channel response for the FDE using soft-decision data symbols based on the log-likelihood ratio (LLR) of extrinsic information provided by a Max-Log-maximum a posteriori probability (MAP) decoder and RS symbols. Computer simulation results show that the threshold based effective path selection in the time domain suppresses the noise components more effectively than a weighted coherent averaging filter in the frequency domain for DDCE as well as for channel estimation with a RS only in terms of the block error rate (BLER) performance level. It is also shown that the DDCE using threshold based effective path selection in the time domain significantly contributes to the decrease in the required average received signal-to-noise power ratio (SNR) for time-multiplexed RSs particularly for a wide transmission bandwidth in DFT-precoded OFDMA.