Segmentation of nonstationary signals using stochastic characteristics of the window variance

Signal or response partitioning (i. e., signal segmentation) is of great interest, e. g., for biomedical research. Signal segmentation, being an essential part of signal processing, may serve as a tool for advanced signal interpretation and data classification. Segmentation of nonstationary signals with a small signal-to-noise ratio is a particulary complicated task. The paper is mainly devoted to exploration of the window variance noise component as a random variable for the proposed signal models. Some stochastic characteristics of the window variance noise components are investigated in accordance with the models. Theoretical findings are consistent with the previously obtained empirical characteristics of the window variance noise component and are supposed to be of potential use for signal segmentation and prediction.