OFDM peak power reduction using metric-based amplitude predistortion

In this paper, we propose a new peak-to-average power ratio (PAPR) reduction technique for OFDM systems. It consists of predistorting the amplitude of a set of symbols per block using a simple metric-based algorithm, and it does not require transmitting any side information to the receiver. The metrics measure how much each symbol contributes to the output signal samples of large magnitude, and the predistorted symbols are selected as those with the largest metrics. The proposed technique can be implemented as a one-shot procedure or as an iterative algorithm. Its performance is investigated using the QPSK signal constellation.