Print Email Facebook Twitter Transistor-Level Statistical Timing Analysis: Solving Random Differential Equations Directly Title Transistor-Level Statistical Timing Analysis: Solving Random Differential Equations Directly Author Tang, Q. Contributor Charbon, E. (promotor) Faculty Electrical Engineering, Mathematics and Computer Science Department Miroelectronics Date 2013-04-02 Abstract In this Ph.D. thesis, a novel non-MC Random differential Equation based Statistical Timing Analysis (RESTA) method is proposed, which considers both process variations and electrical circuit effects, such as multiple input simultaneous switching and crosstalk effects. To make the approach practical for analysis of large circuits at high accuracy, we propose a Simplified Transistor Model (STM) for transistor-level timing analysis. For statistical delay calculation, instead of simulating thousands of times using Monte Carlo methods or simulating at many corners using corner-based methods, RESTA simulates only once, solving random differential equations directly. Crosstalk effects are taken into account based on our proposed Piecewise Linear Delay change curve Model (PLDM) for statistical interconnect delay calculation. Subject statistical timing analysisprocess variationstransistor levelrandom differential equationcrosstalk effect To reference this document use: https://doi.org/10.4233/uuid:c9f651d9-28af-4324-96d7-a2c41afe3529 ISBN 9789461861344 Part of collection Institutional Repository Document type doctoral thesis Rights (c) 2013 Tang, Q. Files PDF thesis_Qin_Tang_print.pdf 6.17 MB Close viewer /islandora/object/uuid:c9f651d9-28af-4324-96d7-a2c41afe3529/datastream/OBJ/view