Research Of Channel Estimation And Equalization Methods For OFDM System Under High Mobilities

Since the large economic and political benefits brought by high-speed railway technology to Japan, more and more countries over the world began to research that technology. High speed mobility of the train brings new challenge to the mobile communication system. Because of the high speed mobility, which leads to terrible Doppler frequency shift, it is difficult for the terminal to demodulate the signal. Furthermore, due to the high mobility, handover becomes frequent, which makes the communication much more difficult.Orthogonal Frequency Division Multiplexing divides the useful spectrum into many orthogonal sub-carriers. Each sub-carrier takes an independent user data. Because of the orthogonality of the data, no interference exists between the adjacent users. This feature of OFDM makes it a high spectral efficiency system as compared to conventional modulation schemes. What’s more, cyclic prefix inserted in the beginning of the OFDM signal makes it robust against intersymbol interference (ISI) and fading caused by multipath propagation.Applying the OFDM to the high speed railway not only is the requirement of the technology, but also is in accordance with the development of the communication technology. However, high mobility leads to terrible Doppler frequency shift, which damages the orthogonality between the sub-carriers. As a result, the inter-carrier interference between the sub-carriers makes the OFDM technology application to the high speed railway difficult.This subject introduces the development of the high speed railway technology at home and abroad firstly. And then, the basic knowledge and features of OFDM systems are presented. On basis of the former introduction, channel estimation methods of OFDM system under high mobilities are studied detailed. Also, we verify the accuracy of estimation results and complexity of the algorithm using MATLAB simulation tools. Due to the inter-carrier interference brought by high mobilities, equalization algorithms are presented in the forth section of the paper. And verifications of them are shown in the sixth section. Furthermore, on basis of the previous research, we propose new channel estimation methods. Improved complex-exponential basis expansion model for channel estimation, combining the Least Square and basis expansion model, combining the linear channel estimation and basis expansion model are presented in the fifth section, respectively. The last method is similar with the former one, but has better spectrum efficiency with similar accuracy.