一种基于PCA的MPSK信号码元速率估计方法

摘要

Aiming at the symbol rate estimation of MPSK signals,this paper investigated the phenomenon that the peak features of cyclic spectrum were corrupted by colored background noise in the case of limited data,and proposed a symbol rate estimation algorithm which combined principal component analysis with cyclic spectrum.PCA transformation restrained the colored background noise in the cyclic spectrum,improved estimation precision,and decreased estimation variance.Simulation result shows that the algorithm acquires fine estimation precision under limited data,and is suitable for MPSK signals with distinct shaped filter coefficients.%针对MPSK信号的码元速率估计问题,研究了有限数据条件下循环谱的谱线特征受到背景色噪声干扰的现象,提出了一种基于主分量分析(PCA)的循环谱特征码元速率估计方法.PCA变换抑制了信号循环谱中的背景色噪声,提高了估计精度,减小了估计方差.仿真表明,该方法在有限数据条件下具有良好的估计性能,适用于不同成形滤波系数的MPSK信号.