Research On Channel Estimation And Signal Detection Technology Of OFDM System In The High-Speed Railway Environment

In the high-speed railway environment, the wireless channel will be time-varying and the OFDM technology will face inter carrier interference; In order to save bandwidth or to adapt to different multi-path environment, considering the OFDM system with the inadequate cyclic prefix, which bring the inter symbol interference of the OFDM system. In order to maximize the elimination of inter carrier interference and inter symbol interference, to achieve accurate data transmission, the paper focuses on OFDM system channel estimation and signal detection in the high-railway environment.The second chapter focuses on the high-speed railway channel model based on basis expansion model. Firstly, we study the typical channel model in high-speed railway environment and the base expansion model and its classification, focusing on the popular complex exponential basis expansion model, the high frequency part of the existing sampling error, we can use the modified complex exponential basis expansion model to make sure that the sampling points fall inside the Doppler spread spectrum and the maximum sampling frequency is equal to the maximum Doppler frequency shift. Then, we study and propose the optimization design method of improved basis dimension and basis function of the modified complex exponential basis expansion model, theoretical analysis and simulation shows that the base dimension mainly depends on the channel maximum doppler frequency shift and signal to noise ratio, When the number of pilots is fixed, the higher the base dimension, the smaller the modeling error, but the accuracy of parameter estimation will be reduced, we need to make a compromise between the two, the basis function will be determine after the base dimension, influence on doppler extended accuracy of spectral representation.The third chapter studies the optimization design methods of comb type pilot in the OFDM system under time-varying channel. Firstly, based on linear minimum mean square error, we study the optimal cluster-type pilot design, from the relations between channel time domain and frequency domain matrix transformation matrix of OFDM system, then use the discrete Karhuen-Loeve basis expansion model to equal the frequency domain matrix transformation matrix, and the analysis of the receiver using linear minimum mean square error estimation equation and the performance criterion, discussed under certain conditions, pilot power, the mean square error optimal comb pilot design requirements;We study the frequency-domain Kronecker delta structure(with each group, one nonzero pilot is placed in the middle with some zero guard on both sides), increasing the amount of information carried by the number of increased nonzero pilot number per cluster. Although the inter-carrier interference has improved, the performance of channel estimation will be improved. Aiming at the high the peak to average ratio in tradition pilot structure, the pilot data is same between clusters, each cluster pilot data with zero correlation zone sequences, throught the method, each cluster is multiplied by a pseudo-random number, to reduce the peak to average ratio.The fourth chapter studies the channel estimation and signal detection algorithm of OFDM system based on comb pilot in fast time-varying channel. Firstly, we propose the channel estimation algorithm for multiple OFDM symbols based on comb type pilot structure. The number of M OFDM symbols are regarded as the basic channel estimation unit, in the period of duration based expansion model only a basic dimension approximation to the floor, the total channel estimation parameter is reduced, and the use of the same pilot measurements to estimate the channel estimation, which bring the actual estimation performance improvement. Next, according to the environment of a large multi-path delay, the time-varying channel under the condition of insufficient cyclic prefix, the channel estimation and signal detection for OFDM system based on basis expansion model, we derived inter-symbol interference cancellation techniques and circular convolution reconstruction technique under the base expansion model from theory, and numerical simulation results demonstrate that the proposed algorithm can achieve better system performance.To sum up, this paper from the high-speed rail channel model, the pilot design and the channel estimation and signal detection, discusses the application of the basis expansion model which has the theoretical research and practical application value.