Abstract
Due to the demand of high data-rate and reliable data transmission over wireless networks, orthogonal frequency-division multiplexing (OFDM) becomes one of the attractive technologies for communication systems. The OFDM systems can offer not only high data-rate services but also robustness to frequency selective fading channels. One of the major drawbacks of OFDM signals is the high peak-to-average power ratio (PAPR). When a high PAPR signal passes through a power amplifier, it may cause nonlinear distortion. Numerous techniques have been proposed to reduce the PAPR of OFDM signals, such as selected mapping (SLM) and constellation extension (CE). The SLM method can provide good performance on PAPR reduction, but it needs to transmit side information to the receiver. The CE method can also provide good PAPR reduction performance by appropriately extending the modulation constellation of input data symbols; however, it will increase the transmit power and the computation complexity at the transmitter. In this thesis, we propose a low-complexity constellation extension scheme to reduce the PAPR of OFDM signals. The proposed method combines the concept of the conventional SLM and CE schemes, where the average increasing power and the bit-error-rate (BER) due to the extended constellations are limited to an acceptable region. By using some low-complexity conversion matrices T, the proposed scheme generates multiple candidate signals and the signal with the lowest PAPR is selected for transmission. Moreover, the proposed scheme does not need to transmit side information. As compared with the conventional SLM and CE schemes, the proposed one achieves close PAPR reduction and BER performances with much lower computational complexity.