This document discusses non-parametric power spectrum estimation methods. It introduces applications of power spectrum estimation such as feature extraction and Wiener filtering. The periodogram method is described as directly calculating the discrete Fourier transform of the data, but it has high variance. Improved methods average multiple periodograms, such as Welch's method, or apply windows to the data or periodograms, like Bartlett's and Blackman-Turkey methods. These methods trade off bias and variance to provide better resolution and convergence to the true power spectrum as the data length increases.