Optimization of LSE and LMMSE Channel Estimation Algorithms based on CIR Samples and Channel Taps

For spectrally efficient transmission over time-varying channels, the use of Adaptive Coding and Modulation (AMC) in wireless OFDM systems requires the estimation of radio channel at the receiver. This paper focuses on the use of time domain channel statistics, mainly concentrating on two schemes: Linear Minimum Mean Square Estimation (LMMSE) and Least Square Estimation (LSE) and their variants. LMMSE performs better than LSE but at the cost of computational complexity. The performance of LSE can be improved by increasing CIR samples and channel taps. To avoid the matrix inversion lemma, the channel matrix can be downsampled or regularized. Theoretical analysis and computer simulations are used for performance and complexity comparisons.