An Investigation On Channel Estimation And Diversity Combining Algorithms In LTE PUSCH

With the operation of the world's first LTE (Long Term Evolution) commercial site, mobile communication is stepping into 4G era. TDD (Time Division Duplex)LTE standard have become increasingly sophisticated, and the research of related products enters a critical stage. In order to meet the demand for high-speed data transmission of 4G, SC-FDMA (Single Carrier-Frequency Division Multiple Access) and 1x2 MU-MIMO techniques have been adopted in LTE Uplink. Because the wireless transmission channel has time-varying, frequency selective fading and time selective fading characteristics, the received signals are often seriously distorted. So, accurate channel estimation and excellent diversity combining are the key techniques guaranteed to a quality of uplink receiver signals.Therefore, this paper will focus on research and design channel estimation and antenna diversity combining techniques for LTE uplink PUSCH (Physical Uplink Shared Channel). The channel estimation method based on reference signals, first estimates channel response in the location of the reference signals, then get the all data location's channel response through interpolation. The LS (Least Square) and the MMSE (Minimum Mean Square Error) channel estimation algorithm is studied in this paper, then a simplified MMSE estimator is proposed for the PUSCH , the simulation results show that the simplified scheme has good performance and low complexity compared with MMSE algorithm. And tow interpolation algorithm, linear interpolation and DFT (Discrete Fourier Transform) interpolation, will be discussed, and simulation result shows that the performance of DFT interpolation algorithm is slightly better than that of the first-order linear interpolation. The antenna diversity combining is implemented after all frequency data equalization. For diversity combining techniques, we mainly focus on the maximum SNR combining, equal gain combining and selection combining techniques, simulation results show that maximum SNR combining scheme has the best performance.