基于AR模型谱估计方法与仿真

Spectrum Estimation Method and Simulation Based on AR Model

功率谱估计在研究信号特性中占有着十分重要的地位,是获取随机信号频谱分布的常用方法.详细阐述了AR模型的基本理论,分析了其实现的Yule-Walker法,协方差法和修正协方差法,Burg递推法.并用MATLAB软件对上述几种谱估计的方法进行一些仿真,比较了其预期的谱估计效果.开展利用AR模型法进行功率谱估计方法的研究,以便在应用中根据实验数据选择合适的算法进行谱估计.

Power spectrum estimation plays an important role in the study of signal characteristics.It is a common method to obtain the spectrum distribution of random signals.In this paper,the basic theory of AR model is described in detail.Then,Yule-walker method,covariance method,modified covariance method and Burg recursion method for AR implementation are analyzed.Finally,several spectral estimation methods mentioned above are simulated with MATLAB software,and the expected spectral estimation results are compared.The research of power spectrum estimation using AR model method is carried out in order to select suitable algorithm for spectrum estimation according to experimental data in application.