Research Of Techniques Of Channel Estimation In Wireless OFDM System Based On Comb-Type Pilots

Orthogonal Frequency Division Multiplexing (OFDM) technique has been considered as one of the cutting edge techniques in 4G mobile communication because it can combat multipath fading and achieve high data rate. Accurate channel estimation is the one of the key techniques for reliable OFDM communication. In OFDM system, receiver must acquire Channel State Information (CSI) to equalize multipath fading. This thesis mainly aims at researching the techniques of channel estimation in wireless OFDM systems.Firstly, the background and development of OFDM technique are introduced, the characteristics of OFDM systems and the techniques of channel estimation are summarized. Then, the characteristics and models of various wireless channel and the basic principles of OFDM system is discussed in detail. Moreover, two primary types of channel estimation which are based on pilot-symbol and decision feedback are analyzed in particular. And improved algorithms of channel estimation are proposed which can be applied in wireless OFDM systems.Further, several common interpolation algorithms are analyzed and simulated based on comb-type pilots LS algorithm. The simulation results show that the performance of filter interpolation is more stable. Meanwhile, given the weak accuracy of LS algorithm, IPF-LS and IDF-LS are proposed which can enhance the accuracy of channel estimation by employing time-domain low-pass filter.Generally, decision feedback channel estimation is only applied non-pilot or block-type pilots OFDM system. A novel decision feedback channel estimation algorithm with double receiver is proposed in this thesis, which can be employed in comb-type pilots OFDM system. Simulation results display that the algorithm functions much better than comb-type pilots LS channel estimation algorithm. Furthermore, IPF-LS and IDF-LS has been applied to improve this proposed algorithm, the improved algorithm shows excellent performance.