Automated state change detection of a Gaussian distributed signal.

摘要

This work presents a novel approach to detect a change in the state of a signal. The hypothesis is that in a given state, the values of a signal vary in a sub-part of a Gaussian distribution. By modeling a signal using a sub-part of the Gaussian distribution, a quantitative value representing the state of the signal is obtained. This value is monitored with time to detect a change in the state of the signal. The proposed algorithm is implemented and tested on data obtained from a related research.