Research Of SCM Channel Modeling And Channel Estimation For LTE-OFDM System

By the end of 2004, the Third Generation Partnership Project (3rd Generation Partnership Project,3GPP) started the Universal Mobile Telecommunications System (Universal Mobile Telecommunications System, UMTS) technology Long Term Evolution (Long Term Evolution, LTE) project to ensure its long-term comparative advantage of wireless standards. Both the multiple antenna (MIMO) technology, which will extend the spatial domain to improve system capacity, and orthogonal frequency division multiplexing (OFDM), which will meet the requirements of the flexibility for spectrum allocation and wide high-speed whit low-cost carrier solutions in LTE standard, are key technologies in the new standard of LTE. Spatial Channel Model (Spatial Channel Model, SCM) is the MIMO channel model for 3GPP-3GPP2 LTE system simulation. For the OFDM technical characteristic as well as wireless channel's fading characteristic, channel estimation technology based on LTE OFDM system has been widely studied in the field. The channel estimation is one of the key technologies to realize the wireless communication, which directly affect the accuracy of demodulating the start information in the receiving end.In this work, the LTE Physical Downlink has been established according to 3GPP TS 36.211 firstly, then the SCM channel model has been established based on sub-paths method according to 3GPP TR 25.996, and the LTE Physical Downlink Shared Channel simulation has been accomplished. In the view of OFDM technical characteristic, pilot-based channel estimation for time-frequency two-dimensional is used in LTE OFDM system. In this paper, The channel estimation is investigated for different pilot densities in frequency-domain and pilot allocation in temporal-domain for low delay spread and high delay spread channels. Channel estimation based on LS (least squares) and MMSE (minimum mean square error) algorithm is studied. Finally an improved algorithm, the dominant channel taps selected (DCT), based on the traditional LS algorithm in temporal-domain is presented. The performance of these schemes have been compared by measuring the mean-squared-error (MSE) or raw bit-error-rate (BER) with QAM modulation on Rayleigh fading channel based on SCM model. The simulation results show that the proposed algorithm can further suppress the noise interference and improve the estimation performance, while controlling the algorithm complexity.