Performance Comparison of LS and LMMSE Channel Estimation Methods for MIMO-OFDM LTE Downlink System

Orthogonal Frequency Division Multiplexing -LTE downlink system provides higher data rates with the use of MIMO-OFDM techniques. Due to the trade-off between complexity and estimation accuracy, choice of channel estimator is important factor for LTE downlink system. For pilot aided type, Least Square(LS) method exhibits lower complexity & requires implicit channel knowledge. But it suffers from Inter Carrier Interference (ICI). On other side, Linear Minimum Mean Square Error (LMMSE) method performs better than LS with higher computational and circuit complexity. Due to this, optimal design of Channel Estimator is area of ongoing research. As LTE Downlink system is based on a MIMO-OFDMA technology, a Cyclic Prefix (CP) is inserted at the beginning of each transmitted OFDM in order to minimize inter-carrier interference (ICI) and inter-symbol interference (ISI). Usually, the length of the inserted CP is defined to be equal to or longer than the channel length. Here we proposed the performance analysis of two estimation methods. Simulation results show that when cyclic prefix length is equal to or longer than channel length, LMMSE has better performance than LS. In other case if cyclic prefix is shorter than channel length, LMMSE performs better only for lower SNR. As SNR increases, LS shows better performance.