*********************************
There is now a CONTENT FREEZE for Mercury while we switch to a new platform. It began on Friday, March 10 at 6pm and will end on Wednesday, March 15 at noon. No new content can be created during this time, but all material in the system as of the beginning of the freeze will be migrated to the new platform, including users and groups. Functionally the new site is identical to the old one. webteam@gatech.edu
*********************************
Title: Behavioral Modeling of Drivers and Oscillators Using Machine Learning
Committee:
Dr. Madhavan Swaminathan, ECE, Chair , Advisor
Dr. Sung-Kyu Lim, ECE
Dr. Arijit Raychowdhury, ECE
Dr. Saibal Mukhopadhyay, ECE
Dr. Suresh Sitaraman, ME
Abstract:
The objective of this dissertation is to develop time-domain behavioral models for I/O drivers and oscillators for fast simulation and IP protection. For oscillators, augmented neural networks (AugNNs) are proposed to capture the oscillatory behavior of fixed-frequency oscillators and VCOs. When output buffer is included as a part of the oscillator circuit, AugNN-based models are developed taking into account the I/O behavior of the oscillator. For tunable drivers with pre-emphasis, state-aware weighting functions are proposed, and the dynamic memory characteristics of the driver’s output stage are captured using recurrent neural networks (RNNs). The behavior of the tunable control parameters is captured. Furthermore, a transition-variational model is discussed for the modeling of I/O drivers under overclocking conditions. The proposed models are compatible with Verilog-A.
