Statistic-based Channel Estimation And ICI Mitigation In Ofdm System Over High Mobility Channel

With the development of high speed train(HST), Doppler Frequency Shift(DFS) becomes an important factor in wireless communication over the high mobility channel, and the wireless channel is frequency-selective and time–varying, which means it is difficult to estimate the channel information. Meanwhile, Orthogonal Frequency Division Multiplexing( OFDM) is widely adopted in the communication system due to high spectral efficiency and high data rate, however, it brings new problems in the high mobility channel. At first, DFS destroys the orthogonality and causes inter-carrier interference(ICI) which degrade the performance of the OFDM system. Secondly, channel varies very fast in the high mobility environent, as a result, a reliable channel estimation method should be investigated to ensure the communication. New technologies are required to solve these problems and they have great significance in wireless communication over high mobility channel.In this paper, a statistics-based channel estimation and ICI mitigation scheme is proposed for the OFDM system over high mobility channel. From the performance of the proposed scheme compared with other existing schemes, it can be seen the proposed scheme has improved the channel estimation accuracy and frequency utilization, and reduced the complexity. The main work of the proposed scheme has the following three parts:Firstly, a new channel model is proposed for high mobility channel based on the Jakes model of Rayleigh channel and mathematical knowledge. Choosing a few number of impulse response to represent the all propagation in order to reduce the number needed to be estimated, and recover the channel propagation in a certain method which minimizes the channel error.Secondly, a method is proposed for ICI mitigation. As the frequency domain channel matrix is banded in the high mobility environment, we develop a receive post processing in the receiver to concentrate the power of the matrix to the diagonal line, in this way, we can reduce the influence of ICI.Thirdly, this paper provides a novel channel estimation scheme based on compressed sensing for the communication in high speed train scene, and a compressed sensing recovery method utilizing the sparsity of the high mobility system is proposed for channel estimation. Numerical results and simulation verify the effectiveness of the proposed schemes.