Researches On Techniques Of Reducing The Peak-to-Average Power Ratio In Multi-Carrier Communication Systems

Multicarrier communication technology is usually applied to the high speed data transmission in the telecommunication system. It can resist frequency selective fading and have high frequency efficiency. Meanwhile, multiple-input-multiple-output (MIMO) system architecture makes it possible to increase system capacity and improve data transmission speed and system transmission quality. Consequently, they become key technologies in the new generation telecommunication system. Multicarrier communication system with MIMO technology can provide wider covering area, better transmission quality and achieve higher data transmission speed and frequency efficiency. However, the high peak-to-average power ratio (PAPR) problem of multicarrier communication technology becomes one of the main drawbacks in its realization. This thesis is interested in this problem.First of all, this thesis introduces characteristics of the multicarrier communication system and current researches on reducing PAPR schemes, and summarizes problems needed to be resolved.Secondly, this thesis focuses on the research of the PAPR statistic characteristic of the multicarrier communication system. The PAPR statistic characteristic of the discrete signal according to Nyquist sampling has attracted much attention and become the theoretic foundation of many schemes. However, the analysis to the continuous signal is not enough. Therefore, this thesis analyzes the random process composed of continuous signals, and has got the closed form of the signals' correlation at any time. Apply the expression to any oversampled situation. We can find that correlation value between sampled points ascends and the influence that the new sampled point acts on PAPR decreases with the increase of the oversampling rate. Moreover, analysis results prove that the independent distribution relationship between sampled points can be satisfied when oversampling rate is twice as high as the Nyquist sample rate. The theoretic probability distribution function of PAPR at this moment can be got.Thirdly, the common scheme utilizing several alternative signals to control PAPR is studied in this thesis. By theoretically analyzing the performance of reducing PAPR with partial transmit sequences (PTS), we finds that it is dependent on the correlation of several alternative signals. And it is decided by subblock partitions and phase factors, while it is unconcerned with the input signal. Furthermore, this thesis proves that the random subblock partition is more favorable than the interleaved and the adjacent subblock partitions. Meanwhile, theoretical results indicate that there is a floor of the quadratic sum of the correlation function among several alternative signals with random subblock partition. When the number of subblocks is decided, the floor ascends with the number of alternative signals, and the system performance improvement tends to be saturated. The thesis proposes to design subblock partitions...