基于MATLAB实现的AR模型功率谱估计

摘要

现代功率谱估计和经典功率谱估计是两种常用的功率谱估计,同时也是分析随机信号的常用方法.本文详细介绍了现代功率谱估计中有关AR模型参数的功率谱估计,具体包括自相关算法、Burg算法、协方差算法以及改进的协方差算法,在此基础上又对四种不同功率谱估计方法的性能指标进行了比较分析.在MATLAB仿真软件平台上对AR模型参数的四种不同功率谱估计算法进行了仿真,同时对功率谱估计结果进行了分析比较并得到了预期的谱估计效果.最后从实际应用角度出发讨论了AR模型参数不同功率谱估计算法的特点,以便在应用中能够选择合适的功率谱估计算法.%Power spectrum estimation can be divided into modern spectral estimation and classical spectral estimation, it is an method for analyzing random signal. It describes the autocorrelation algorithm, Burg algorithm, covariance algorithm and improved covariance algorithm of AR model parameters in Modern Spectral Estimation , and their perform indicators are analyzed in this paper. These methods for AR model parameter algorithm are implemented by MATLAB , the results of simulation are analyzed and get a better spectrum estimation. Moreover , the advantages and disadvantages of these methods are discussed from the experimental view so as to make reasonable selection in practical work.