Ofdm System Based On The Linear Fm Signal Synchronization And Channel Estimation And Equalization,

With the development of modern society, the researches on the fourth generation (4G) wireless communication system have been started. The 4G system is expected to provide more categories of services, more reliable data transmission and greater transmission rate. However, the environment of wireless communication is very complicated, such as the decrease of average signal power with the increase of transmission distance, the frequency selective fading caused by the multi-path channel and the time selective fading caused by the Doppler effect. All of these block the performance improvements of the wireless communication systems.As a parallel transmission technique with multi-carriers, Orthogonal Frequency Division Multiplexing (OFDM) is very suitable for high data rate transmission because it has the advantages of being robust to frequency selective fading, high spectrum efficiency and simple equalizer. But compared with the single carrier systems, OFDM is sensitive to synchronization errors and has relative large peak to average power ratio (PAPR), which are the major problems should be considered in OFDM systems. The Multiple Input Multiple Output (MIMO) technique can considerably improve the system capacity and spectrum efficiency without the requirement for additional bandwidth. The combination of MIMO and OFDM has become one of the key technologies of the next generation wireless communication systems.This thesis mainly focuses on the techniques of synchronization, channel estimation and equalization in OFDM and MIMO OFDM systems, which are the basic components of the wireless communication systems.Firstly, after the descriptions of the characteristics of wireless channels and the fundamentals of OFDM systems, the effects to OFDM systems of timing errors, carrier frequency offsets and sampling frequency offsets are discussed. The existing solutions are also given. Then, the definition, properties and discrete computation methods of the fractional Fourier transform (FRFT) are introduced. For timing and carrier frequency synchronization in OFDM systems, the possibility of using the linear frequency modulation (LFM) signals and FRFT and the possible performance improvement that may be obtained are analyzed in details.Secondly, two training sequences with different structures are designed which are composed by multiple LFM signals with different modulation frequencies. Correspondingly, two different joint timing and carrier frequency synchronization algorithms based on FRFT are proposed by using the two training sequences that are superimposed on the data signals. The proposed algorithms are then generalized for the distributed and centralized MIMO OFDM systems respectively. Compared with the existing methods, theoretical analysis and simulation results show that the proposed algorithms have better accuracy of timing synchronization and the maximum possible carrier frequency offset that can be estimated is near half of the useful bandwidth no matter in OFDM systems or in MIMO OFDM systems. Furthermore, the transmission of the training sequences does not reduce the system spectrum efficiency because the training sequences are superimposed on the data signals.Thirdly, the structures of the two training sequences are modified to retain the cyclic properties of the transmitted signals. According to the modified training sequences, the channel estimation and equalization methods for OFDM and MIMO OFDM systems are proposed. Since the modified training sequences can also be used to realize the joint timing and carrier frequency synchronization algorithms based on FRFT, a complete OFDM or MIMO OFDM receiver is obtained.Fourthly, the MIMO channel equalization method based on autocorrelation matching (AM) principle is studied and modified for MIMO OFDM systems. The prefixes with different structures are inserted before the data signals corresponding to different transmit antennas so that the autocorrelations of the transmitted signals satisfy the autocorrelation independent condition. Theoretical analysis and simulation results indicate that the complexity of the system is reduced and the bit error rate performance is improved with the modifications compared with the traditional AM method.At last, a conclusion is drawn and the future of wireless communication is envisioned.