Performance analysis of pilot aided channel estimation methods for LTE system in time-selective channels

This paper deals with the performance analysis of channel estimation methods for LTE downlink system over time varying mobile environments. The analysis of channel estimation in the presence of interference is also done. Least square frequency domain (LS_Freq), least square time domain (LS_Time), maximum likelihood (ML) and minimum mean square error estimation (MMSE) techniques are focused for different bandwidth by varying the number of resource blocks used in OFDM structure of LTE system. As number of pilots increases with increasing bandwidth, performance of the estimator increases. Computer simulations are used in the comparisons. Linear Interpolation is performed to get the channel estimation of data symbols. Theoretical results are derived for LS_Time and ML techniques. MMSE performs well at low SNR. LS_Time and ML performs well at high SNR as they use the knowledge of number of channel taps. We achieve interference mitigation using channel estimation using repeated interference features obtained from the two types of pilot symbols used with different pilot density. This technique not only makes the channel estimation robust to the time-selectivity of the channel but also reduces the number of pilot subcarriers needed to estimate the channel.