基于四阶混合累积量的衰减指数模型估计

Estimation of Damped Exponential Model Using Fourth Order Mixed Cumulants

定义了衰减指数模型的四阶混合累积量,给出了单样本有限长数据条件下的样本估计,提出了高斯色噪声条件下估计模型参数的两种有效方法。仿真实验结果表明:四阶混合累积量能明显提高KTProny、MP和ESPR IT方法在高斯色噪声条件下的估计性能,且后两种方法的估计性能优于第一种方法。

The FOMC （fourth order mixed cumulants） of damped exponential model is presented when only a single data record with finite length is available in this paper. Two efficient approaches for the estimation of the parameters of exponentially damped sinusoids are introduced. Monte Carlo simulations demonstrate that the fourth order mixed cumulants can significantly improve the performance of KTProny、MatrixPencil and ESPRIT methods in the presence of additive colored Gaussian noise and the last two methods have better accuracy.