Research Communication | Open Access
Volume 2021 | Communication ID 229
Volume 2021 | Communication ID 229
| Online Abrupt Changes Detection Using Convex Minimization |
Zakariae Drabech, Mohammed Douimi, El Moukhtar Zemmouri
Received |
Accepted |
Published |
January 31, 2021 |
February 15, 2021 |
March 15, 2021 |
Abstract: The detection of abrupt changes in the properties of a data sequence have a wide range of applications such as in robotics, in finance, and data mining. In this paper, we present a novel online detection of abrupt changes in a signal of independent normal observations. The key idea of our algorithm is to estimate the unknown post-change parameters of the normal distributions, by modeling a conditional probability of observations. This modeling can be then reduced to the resolution of a convex minimization problem of energy function at each time instant. Numerical results shows that our ...