Search Results

Since 2000, avionics is facing several changes, mostly driven by technological improvements in the electronics industry and innovation requirements from aircraft manufacturers. First, it has progressively lost its technological leadership over innovation processes. Second, the explosion of the electronics consumer industry has contributed to shorten even more its technology life cycles, and promoted the use of COTS. Third, the increasing complexity of avionics systems, which integrate more and more functions, have encouraged new players to enter the market. The aim of this article is to analyze how technological changes can affect the competitiveness of avionics firms. We refer to criticality levels as a determinant of the market competitiveness. Certification processes and costs could stop new comers to bring innovations from the consumer electronics industry and protects traditional players.

For more than 40 years, Gordon Moore's experimental law has been predicting the evolution of the number of transistors in integrated circuits, thereby guiding electronics developments. Until last years, this evolution did not have any measurable impact on components' quality; but the trend is beginning to reverse. This paper is addressing the impact of scaling on the reliability of integrated circuits. It is analyzing - from both qualitative and quantitative point of view - the behavior of Deep Sub-Micron technologies in terms of robustness and reliability. It is particularly focusing on three basics of safety analyses for aeronautical systems: failure rates, lifetimes and atmospheric radiations' susceptibility.

The changes brought by the increasing integration density and the new technological trends have pushed the reliability at its limit. Safety analysis for critical system such as embedded electronics for avionics systems needs to take into account these changes. In this paper, we present the consequences on the deep sub-micron (DSM) CMOS devices concerning their single event effect (SEE) sensitivity. We also propose a new modeling method in order to address these issues.