根据维纳-辛钦定理，广义平稳随机过程的功率谱是自相关函数值。但对于许多实际应用问题，可资利用的观测数据往往是有限的，所以要准确计算功率谱通常是不可能的。比较合理的目标是设法得出功率谱的一个好的估计值，这就是功率谱估计。本文通过分析经典功率谱估计和现代功率谱估计的多种方法，比较各种方法的优缺点。-According to Wiener- Xinqin theorem, wide-sense stationary random process is the power spectrum of the autocorrelation function value. But for many practical applications, Keziliyong observational data are often limited, so to accurately calculate the power spectrum is usually not possible. A more reasonable goal is to try to come to a good power spectrum estimation, this is the power spectrum estimation. This paper analyzes the power spectrum estimation and modern classical power spectrum estimation of a variety of methods to compare the advantages and disadvantages of various methods.