基于曲线拟合的多项式相位信号参数估计

Parameter estimation of polynimial phase signals based on curve fitting

研究了多项式相位信号的参数估计问题。针对在分析高阶多项式相位信号时,维格纳-威利分布的交叉项使得时频分布图变得难以解释的问题,采用伪维格纳-威利变换减少交叉项,并且从伪维格纳-威利分布提取出多项式相位信号的瞬时频率曲线,利用曲线拟合的方法拟合瞬时频率曲线,从而实现了多项式相位信号参数估计。对比了最小二乘曲线拟合和切比雪夫曲线拟合用于多项式相位信号参数估计的不同。最后通过仿真验证了曲线拟合用于多项式相位参数估计的有效性。

A parameter estimation method of polynomial phase signals was studied. Due to cross-terms in the Wigner- Ville Distribution （WVD） of the ploynomial phase signals, the figure of time-frequency distribution becomes difficult to understand. In this paper, cross-term attenuation was obtained by using pseudo Wigner-Ville transformation, then the instantaneous frequency curve could be extracted from pseudo Wigner-ViUe distribution. The curve fitting algorithm was used to estimate parameters for the polynomial phase signals. Least squares curve fitting was compared with Chebyshev curve fitting. Simulation test shows the ettlciency of this method.