SIGNATURE ANALYSIS FOR MEMS PSEUDORANDOM TESTING USING NEURAL NETWORKS

Lukáš Kupka, Emmanuel Simeu, Haralampos-G. Stratigopoulos, Libor Rufer, Salvador Mir, Olga Tumová

Abstract

The aim of this work is to develop a lowoverhead, low-cost built-in test for Micro Electro Mechanical Systems (MEMS). The proposed method relies on processing the Impulse Response (IR) through trained neural networks, in order to predict a set of MEMS performances, which are otherwise very expensive to measure using the conventional test approach. The use of neural networks allows us to employ a low-dimensional IR signature, which results in a compact built-in test. A MEMS structure combining electro-thermal excitation and piezoresistive sensing was chosen as our case study. A behavioral model of this structure was built using Matlab for the purpose of the experiment. The results demonstrate that the neural network predictions are in excellent agreement with the simulation results of the behavioral model.

Keywords

MEMS testing, neural networks, feature selection

Download

IMEKO-TC1-TC7-2008-042.pdf

DOI

-

IMEKO TC

TC1 - Education and Training in Measurement and Instrumentation

Back to the proceedings