Study On Channel Estimation In OFDM/TDM System

Orthogonal Frequency Division Multiplexing (OFDM) is one of the most promising wireless communication techniques, which is robust to ISI (inter-symbol interference) and has been widely used in wideband wireless communication area. However, the high peak-to-average-power radio (PAPR) has been the biggest problem of OFDM system. To alleviate the PAPR problem of the conventional OFDM, OFDM combined with time division multiplexing (OFDM/TDM), also called generalized OFDM (GOFDM) was proposed. By using the time division multiplexing technology for processing signals, OFDM/TDM system greatly reduces the PAPR as well as keeps the advantages of conventional OFDM including high bandwidth efficiency, robustness against multi-path fading channel, easy hardware implementation and so on.Channel estimation is one of the key techniques in wireless communication area as well as a necessary technique in OFDM/TDM system. According to the time-selectivity and frequency-selectivity characteristics of wireless channel in mobile communication, in order to improve communication efficiency and quality, it is necessary to make a real-time channel estimation of wireless channel to obtain the accurate channel state information (CSI) and realize accurate coherent demodulation. By analyzing and summarizing the related references, the thesis mainly studies the time-varying channel estimation in OFDM/TDM system. The main contents are displayed as follows:1) Analyzing and summarizing the basic principles, structures, advantages and disadvantages of conventional OFDM and OFDM/TDM system. Based on the analysis and simulations of the PAPR and spectrum efficiency in the two systems, results show that OFDM/TDM system can highly reduce the PAPR while has a relative lower spectrum efficiency.2) Studying the principle of frequency domain equalization and the process of channel estimation in OFDM/TDM system. Based on the factors affecting the equalization algorithms, we study the equalization criterion based on zero forcing (ZF) and minimum mean square error (MMSE) and analyze the characteristics of the two algorithms. Simulation results show that the equalization technique based on MMSE achieves a better performance than that of ZF but with a much higher computational complexity.3) According to the lack of tracking ability against fast fading channel of the existing pilot structures and channel estimation algorithm in OFDM/TDM system, a method of time multiplexing pilot (TDM Pilot) combined with an interpolation algorithm is proposed in the thesis. By using the channel transfer functions estimated by two adjacent TDM pilots, the channel state information between the pilots are obtained by the interpolation algorithm in this method. Combined with frequency domain equalization based on minimum mean square error (MMSE-FDE), the method can significantly improve the system performance in a fast fading channel. Compared with the original methods, simulation results shows that the method proposed in the thesis can achieve a better performance, which is not only capable of dealing with slow fading channel, but also has a better tracking ability against fast fading channel with the same data transmission efficiency.