Talks and Poster Presentations (with Proceedings-Entry):
R. Hasani, B. Kulnik, D. Haerle, R. Grosu:
"Artificial Intelligence Solutions for Verification of Analog and Mixed-Signal Smart Power Systems";
Talk: 9th International Workshop on Frontiers in Analog CAD at ASYNC 2018,
- 2018-05-17; in: "Proceedings of the 9th International Workshop on Frontiers in Analog CAD",
In this work, we explore novel artificial intelligence (AI) solutions for the verification of analog and mixed-signal (AMS) complex integrated circuits (IC). The main target is to improve the time performance and the accuracy of the verification process of AMS circuits. In this regard, we develop numerous neural network models for AMS ICs, and utilize such models into the verification process. We show that such methods can result in gaining up to two orders of magnitude speed-up in the fault-injection simulations of ICs. In terms of developmental-efficiency and automation performance, our AI modeling suite is superior over existing methods.
Created from the Publication Database of the Vienna University of Technology.