Abstract: In the process of reducing the Peak to Average Power Ratio（PAPR） of the Underwater Acoustic（UWA） Orthogonal Frequency Division Multiplexing（OFDM） systems, the traditional clipping methods are used with high noise for the High Power Amplifier（HPA） working in the linear region effectively. In order to decrease the noise, a novel algorithm with low complexity is proposed. Different clipping values are used to clip different amplitude of signals. And then the above clipping results are recorded by a vector for the recovery of the original amplitude and the clipping noise is decreased. Theoretical analysis and simulation results show that the algorithm can restrain the clipping noise effectively with a low computation and make the amplifier work properly.

Key words: Underwater Acoustic（UWA） communication, Orthogonal Frequency Division Multiplexing（OFDM）, Peak to Average Power Ratio（PAPR）, clipping noise

摘要： 为了降低正交频分复用（OFDM）水声（UWA）通信系统的峰均比（PAPR），使放大器能工作在线性有效区，传统的限幅方法通常会引入较大的限幅噪声。针对这一问题，提出了一种低复杂度的降低限幅噪声的算法。该算法采用不同的限幅门限对不同的信号幅度作限幅处理，同时将处理结果以限幅向量的形式传送至接收端，以准确恢复相应被限数据的原始幅度，从而降低限幅噪声。通过理论分析和计算机仿真，结果表明该算法在少量增加系统运算复杂度的前提下，能使线性放大器工作在有效范围内，而且能够有效抑制系统的限幅噪声。

关键词: 水声通信, 正交频分复用, 峰均比, 限幅噪声