Research On Peak-to-Average Power Ratio Reduction Schemes In OFDM System

Owing to the high spectral efficiency and the immunity to multipath fading, orthogonal frequency division multiplexing (OFDM) has become an effective technique for high-rate-data transmission and has been adopted in many fields of wireless communications, such as digital audio broadcasting (DAB), digital video broadcasting (DVB), wireless local area networks (WLAN), and so on. However, one of the major drawbacks of OFDM system is the high peak-to-average power ratio (PAPR), and it brings on OFDM signal distortion in the nonlinear region of high power amplifier (HPA) and the signal distortion induces the degradation of bit error rate (BER). Thus, PAPR problem becomes an obstacle which confines the practical application of OFDM technique.Recently, how to effectively reduce the PAPR in OFDM system has become an urgent problem. Some valuable schemes for PAPR reduction have been proposed, such as methods based on clipping, algorithoms by the use of block coding and probabilistic schemes. But each of these schemes has its own shortcomings. Clipping is the simplest scheme which clips the OFDM signal to limit the value of PAPR below a threshold level, but it causes signal distortion and big out-of-band radiation. Block coding is another scheme for PAPR reduction which can achieve good PAPR reduction performance, however, there are not enough codewords for a large number of subcarriers. Probabilistic schemes use different phase weighting factor sequences to cope with OFDM signals, and then the one with the minimum PAPR is selected for transmission, such as partial transmit sequences (PTS) and selected mapping (SLM); this category of schemes are very effective for PAPR reduction without any signal distortion, but the main drawback is the large computational complexity. Many scholars have presented some low computational complexity schemes, all of which are at the cost of loss of PAPR reduction performance.This thesis focuses on the probabilistic schemes for PAPR reduction, including PTS and SLM. The common thing in the two schemes is to find an optimal group of phase weighting factors to cope with the OFDM signal, and then obtain the optimal PAPR reduction performance. However, because of the randomness of choosing the values of phase weighting factors and the large number of combination of these possible choices, this searching method causes high computational complexity. Therefore, to deal with the above problem, a sub-optimum PTS (sub-OPTS) and an adaptive multistage SLM (AMSLM) are proposed.Sub-OPTS scheme is based on the published reduced computational complexity method and uses the linear property of inverse discrete Fourier transform (IDFT) to improve the performance for PAPR reduction, which is degraded due to the reduction in computational complexity, without any increase of the number of complex multiplications. Compared with original PTS, i.e., optimum PTS (OPTS), proposed sub-OPTS can achieve almost the same PAPR reduction performance while computational complexity is dramatically reduced.Based on original SLM, proposed AMSLM scheme divides the whole optimization process into the limited number of stages, and then by the use of different phase weighting factor set in each stage to reduce the correlation of candidate signals from different stages, PAPR reduction performance is improved a lot. Meanwhile, the computational complexity of our proposed AMSLM is further reduced by choosing a proper threshold. Compared with original SLM, our proposed AMSLM scheme can obtain almost the same or better PAPR reduction performance while computational complexity is clearly reduced.