Researches And Realizations On Methods Of Reducing The PAPR Based On FFT Properties In OFDM Systems

In Orthogonal Frequency Division Multiplexing (OFDM) systems, the output signals are the superposition of multiple sub-channel signals. Compared with the conventional signals carrier transmission methods, OFDM has the drawback of high peak-to-average power ratio (PAPR). The high PAPR of output signal causes the output signal of OFDM system has large magnitude fluctuation and also leads to increase the dynamic range of amplifier, Digital-to-Analog converter (D/A) and Analog-to-Digital converter (A/D). The results will reduce the efficiency power of radio frequency amplifier and increase the cost of system. If the range of amplifier can not fit the change of the signal, it will incur signal distortion and destroy the orthogonality of sub-channels.According the problems mentioned above, this thesis focuses on the reducing the PAPR methods design for OFDM systems. The main works this thesis are generalized as follows:First of all, this thesis introduces characteristics of the OFDM systems and current researches on PAPR reduction methods, summarizes problems needed to be resolved, and points the main works and innovations of this thesis. Secondly, the thesis focuses on OFDM system model, and illustrates the impact of oversampling rate to the power spectral density of OFDM system. And then the thesis makes theoretical analysis and mathematical derivation on the PAPR, and clarifies the definition, statistical distribution and influence of PAPR. These provide an effective theoretical base for the latter sections. The chapter 3th and chapter 4th are the focus of this thesis.In Partial Transmit Sequence (PTS), the search complexity increases exponentially with the number of sub-blocks. On the third part of this thesis, an improved method with low complexity is proposed. In our improved method, a cyclic shift of the fast Fourier Transform (FFT) in the time domain is employed to improve the PAPR reduction performance.Furthermore, phase factors are optimized only for latter half of candidates, so as to reduce the complexity. The analysis and simulation results show that, in the case of the same number of sub-blocks and phase weighting factors, the proposed method is almost the same as the performance compared with PTS method while the computational complexity can be effectively reduced.And then improved methods with more PAPR reduction performance are proposed.The methods make a better tradeoff between computational complexity and PAPR reduction performance. In our methods, a cyclic shift of the FFT in the time domain is employed to improve the PAPR reduction performance. In the first place, flipping improved partial transmit sequence method (Flipping IPTS) is proposed which use of iterative flipping algorithm to reduce computational complexity. The search complexity of the iterative flipping IPTS increases linearly with the number of sub-blocks. In the second place, threshold improved partial transmit sequence method (Threshold IPTS) is proposed, a threshold is introduced to reduce the complexity in the method. The Matlab simulation results show that the proposed two IPTS methods achieve better performance than PTS method with comparable computational complexity.In the conventional Selected Mapping (SLM), the computational complexity increases in proportion to the number of phase sequences and sub-blocks. On the fourth part of this thesis, we propose improved SLM methods which considerably reduce the number of IFFT blocks with keeping the high PAPR reduction performance compared with the conventional SLM scheme.The proposed methods make use of generated phase sequences and the circular shift property, the linear property of FFT and combination of the two properties to construct three transformations for generating more alternative candidate OFDM signals without increasing IFFT operations.The analysis and simulation results show that improved methods can improve system performance effectually.