This paper addresses the problem of optimizing metalization patterns of back-end connections for the power-MOSFET based driver since the back-end connections tend to dominate the on-resistance Ron of the driver. We propose a heuristic algorithm to seek for better geometric shapes for the patterns targeting at minimizing Ron and at balancing the current distribution. In order to speed up the analysis, the equivalent resistance network of the driver is modified by inserting ideal switches to avoid repeatedly inverting the admittance matrix. With the behavioral model of the ideal switch, we can significantly accelerate the optimization. Simulation on three drivers from industrial TEG data demonstrates that our algorithm can reduce Ron effectively by shaping metals appropriately within a given routing area.
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Bo YANG, Shigetoshi NAKATAKE, "Fast Shape Optimization of Metalization Patterns for Power-MOSFET Based Driver" in IEICE TRANSACTIONS on Fundamentals,
vol. E92-A, no. 12, pp. 3052-3060, December 2009, doi: 10.1587/transfun.E92.A.3052.
Abstract: This paper addresses the problem of optimizing metalization patterns of back-end connections for the power-MOSFET based driver since the back-end connections tend to dominate the on-resistance Ron of the driver. We propose a heuristic algorithm to seek for better geometric shapes for the patterns targeting at minimizing Ron and at balancing the current distribution. In order to speed up the analysis, the equivalent resistance network of the driver is modified by inserting ideal switches to avoid repeatedly inverting the admittance matrix. With the behavioral model of the ideal switch, we can significantly accelerate the optimization. Simulation on three drivers from industrial TEG data demonstrates that our algorithm can reduce Ron effectively by shaping metals appropriately within a given routing area.
URL: https://global.ieice.org/en_transactions/fundamentals/10.1587/transfun.E92.A.3052/_p
Copy
@ARTICLE{e92-a_12_3052,
author={Bo YANG, Shigetoshi NAKATAKE, },
journal={IEICE TRANSACTIONS on Fundamentals},
title={Fast Shape Optimization of Metalization Patterns for Power-MOSFET Based Driver},
year={2009},
volume={E92-A},
number={12},
pages={3052-3060},
abstract={This paper addresses the problem of optimizing metalization patterns of back-end connections for the power-MOSFET based driver since the back-end connections tend to dominate the on-resistance Ron of the driver. We propose a heuristic algorithm to seek for better geometric shapes for the patterns targeting at minimizing Ron and at balancing the current distribution. In order to speed up the analysis, the equivalent resistance network of the driver is modified by inserting ideal switches to avoid repeatedly inverting the admittance matrix. With the behavioral model of the ideal switch, we can significantly accelerate the optimization. Simulation on three drivers from industrial TEG data demonstrates that our algorithm can reduce Ron effectively by shaping metals appropriately within a given routing area.},
keywords={},
doi={10.1587/transfun.E92.A.3052},
ISSN={1745-1337},
month={December},}
Copy
TY - JOUR
TI - Fast Shape Optimization of Metalization Patterns for Power-MOSFET Based Driver
T2 - IEICE TRANSACTIONS on Fundamentals
SP - 3052
EP - 3060
AU - Bo YANG
AU - Shigetoshi NAKATAKE
PY - 2009
DO - 10.1587/transfun.E92.A.3052
JO - IEICE TRANSACTIONS on Fundamentals
SN - 1745-1337
VL - E92-A
IS - 12
JA - IEICE TRANSACTIONS on Fundamentals
Y1 - December 2009
AB - This paper addresses the problem of optimizing metalization patterns of back-end connections for the power-MOSFET based driver since the back-end connections tend to dominate the on-resistance Ron of the driver. We propose a heuristic algorithm to seek for better geometric shapes for the patterns targeting at minimizing Ron and at balancing the current distribution. In order to speed up the analysis, the equivalent resistance network of the driver is modified by inserting ideal switches to avoid repeatedly inverting the admittance matrix. With the behavioral model of the ideal switch, we can significantly accelerate the optimization. Simulation on three drivers from industrial TEG data demonstrates that our algorithm can reduce Ron effectively by shaping metals appropriately within a given routing area.
ER -