Title

Kalman Filtering Based On the Maximum Correntropy Criterion in the Presence of Non-Gaussian Noise

Document Type

Conference Proceeding

Publication Date

2016

Publication Title

2016 Annual Conference on Information Science and Systems (CISS)

Abstract

Power devices have to withstand fast thermal cycling in automotive applications. In order to guarantee the reliability in these applications, a detailed understanding of the degradation mechanisms is required. One of these mechanisms is inter-layer-dielectric cracking, caused by progressive plastic deformation of metal lines. In a previous publication we have shown that DMOS transistors, with different active areas, which operate in similar conditions, have dissimilar reliabilities. This cannot be explained by state-of-the-art methods, which estimate reliability from the peak junction temperature. In this paper, extended measurement results of a DMOS transistor will be analyzed with the aid of electro-thermal and thermo-mechanical simulations and a new approach for reliability estimation will be proposed.

Original Citation

R. Izanloo, S. A. Fakoorian, H. S. Yazdi and D. Simon, "Kalman filtering based on the maximum correntropy criterion in the presence of non-gaussian noise," in 2016 Annual Conference on Information Science and Systems (CISS), 2016, pp. 500-505.

DOI

10.1109/TDMR.2016.2583663

Volume

99