周期图法功率谱估计及其改进算法的研究

Research on power spectrum estimation and improved algorithm of periodic graph method

功率谱估计在随机信号处理领域中发挥了非常重要的作用,有着十分广泛的应用。以研究周期图法谱估计为目的,主要描述周期图法谱估计的算法与性能,并基于MATLAB平台,从算法的原理特性出发,对周期图法谱估计的原理进行了研究。同时给出了两种改进的周期图法,即Bartlett法和Welch法谱估计,采用算法对比分析的方式进行了仿真比较,并说明了3种功率谱估计方法的优缺点。通过应用实例对比说明Welch法通过选取适当的分段长度,可以较好地改善周期图法的性能。

Power spectrum estimation plays a very important role in the field of random signal processing and has a wide range of applications.The algorithm and performance of periodic spectrum estimation is studied.Based on MATLAB platform,the principle of periodic spectrum estimation is studied based on the principle of the algorithm.At the same time,two improved periodic graph methods,namely Bartlett method and Welch spectral estimation,are presented.The simulation comparison is carried out by means of algorithm comparison analysis,and the advantages and disadvantages of the three power spectrum estimation methods are illustrated.By comparing the application examples,the Welch method can improve the performance of the periodogram method by selecting the appropriate segment length.