Abstract: Because of high spectrum efficiency and robustness to multi-path delay, Orthogonal Frequency Division Multiplexing (OFDM) has been focused on in communication research. But the disadvantage of High Peak-to-Average Power Ratio (PAPR) in OFDM technology is one of the main obstacles to limit its wide applications. Various techniques on reducing the PAPR are summarized in this paper firstly, and the definition and distribution of PAPR are introduced. The basic principle of Selected Mapping (SLM) is analyzed. Then the algorithm of using Riemann Sequences to reduce the PAPR is proposed. The main idea of the algorithm is to select row vectors of the normalized Riemann matrix as phase vectors. Some results are obtained by Monte Carlo simulations, and it is proved that this algorithm has better performance for reducing the PAPR of OFDM signals. The computational complexity is reduced, and the bandwidth efficiency is improved.