Researches On New Techniques Of Signal Optimization In OFDM System

Orthogonal Frequency Division Multiplexing (OFDM) is one of the promising techniques for future wireless communication systems, since it is with high spectrum efficiency and robustness to the frequency selective fading environments. However, OFDM has the drawback of high peak-to-average power ratio (PAPR) . In this case, when OFDM signals with high peak values pass through high power amplifier, nonlinear distortions will be introduced so as to degrade the transmission performance. According to the problems methioned above, this dissertation focuses on PAPR reduction techniques for OFDM systems, so as to design the optimal transmission signal.Firstly, the dissertation introduces the features of OFDM systems and the research status of PAPR reduction techniques, summarizes the fundamental problems needed to be solved, and points the main research contents of this dissertation.Secondly, the dissertation makes a theoretical analysis and mathematic calculation on the PAPR of OFDM signals, and clarifies the definition, the distribution and the models of PAPR, so as to make a theoretical base for the latter sections.Partial transmit sequence (PTS) is a class of distortless method for effectively controlling the PAPR of OFDM signals. However, it has the drawback of high computational complexity. On the third part of the dissertation, an improved method with low complexity is proposed. In the improved method, the correlationships among the candidate signals generated from PTS are explored and utilized for simplifying the compuatation for candidate signals. Theoretical analysis and simulation results show that the proposed method can effectively reduce the computational complexity without loss of PAPR reduction performance. Furthermore, an improved method with more PAPR reduction performance is proposed. In the proposed method, the processing for high PAPR signals is enhanced while the processing for low PAPR signlas is simplified, so as to make a better tadeoff between PAPR reduction performance and computational complexity. Another drawback of PTS is the desire of extra side information so as to influence the spectral efficiency. To solve this problem, an improved method with less side information amount is proposed.Clipping is a popular method to control the PAPR with distortions, but it introduces additional clipping noise, so as to distort the data and pilot symbols and degrade the system performance. However, most of current reserches do not consider the influence to pilot-aided cannel estimation by clipping noise. On the fourth part of the dissertation, a new method is proposed to improve the channel estimation performance in clipped OFDM systems. In the proposed method, the data and pilot symbols are divided for individual processing, and the clipping noise on the positions of pilot symbols is filtered directly. In this case, the channel estimation at the receiver is improved, and the system performance of clipped OFDM systems is also enhanced. Based on the new method, an iterative method is proposed to further the channel estimation performance. Meanwhile, for the proposed method, the improvement of pilot-aided carrier frequency offset estimation performance is also considered. Furthermore, for traditional iterative clipping and filtering method, an improved method with more PAPR reduction is proposed. The new method can achieve better PAPR reduction performance on the same number of iterations and complexity as traditional method.Finally, the dissertation investigates the PAPR reduction of multiple access combined OFDM systems. For orthogonal frequency division multiplexing access (OFDMA) system, the PAPR of OFDMA signals with different subcarrier allocation styles is compared, and a low complexity PAPR reduction method for interleaved OFDMA is proposed. For interleave division multiple access (IDMA) combined OFDM system, the PAPR of the transmitted signal is analyzed, and a PAPR reduction method based on variable interleaver designing is proposed.