Research On The Peak-to-Average Power Ratio For OFDM Communication System

As the current rapid expansion in radio communication, services are also a rapidly changing field, which require much higher speed to transmit data. The conventional modulation schemes with single carrier are hard to equalize received signals. OFDM(Orthogonal Frequency Division Multiplexing) technology with good performance on anti multi-path interference and inter-symbol interference is more and more popular. The principle of OFDM technology is to split the entire bandwidth into many orthogonal subchannels, which are overlapped between adjacency, in order to effectively use the whole bandwidth. However, an inherent drawback of OFDM signals is their high Peak-to-Average Power Ratio(PAPR), which is easy to generate intermodulation distortion due to nonlinearity of the high power amplifier. The distortion severely deteriorates the performance of OFDM system. So, we have to solve the problem of high PAPR in OFDM system.In the dissertation, the OFDM system is introduced, including some fundamental theories and key technologies. According to the PAPR problem in OFDM system, the PAPR reduction methods are studied. These PAPR reduction methods mainly include pre-distortion technology, coding technology and probability technology. After deep studying these methods, a novel and effective algorithm is proposed to reduce PAPR of OFDM signals. This proposed algorithm can effectively reduce PAPR with a group of appropriate weights on both amplitude and phase fixed on the subcarriers. Through theoretical analysis, the iterative formula for the PAPR reduction weights is abstracted from this dissertation. The convergence probability of the algorithm is proved, and the convergence limit is given out. Theoretical analysis and computer simulation analyze the impact of the amplitude of the weights on BER and PAPR. Choosing appropriate weights can get good performances in both PAPR and BER. This algorithm can get good PAPR reduction performance as well as satisfactory BER performance, and needs less IFFT operations, which make it very valuable.