Fitzpatrick Institute for Photonics Fitzpatrick Institute for Photonics
Pratt School of Engineering
Duke University

 HOME > pratt > FIP    Search Help Login pdf version printable version 

Publications [#66449] of Nan M. Jokerst

Papers Published

  1. Poddar, Ravi and Moon, Emily M. and Brooke, Martin A. and Jokerst, Nan Marie, Accurate, rapid, high frequency empirically based predictive modeling of arbitrary geometry planar resistive passive devices, IEEE Transactions on Components, Packaging, and Manufacturing Technology Part B: Advanced Packaging, vol. 21 no. 2 (1998), pp. 177 - 183 [96.673706]
    (last updated on 2007/04/16)

    Abstract:
    A novel technique is presented for the high speed, accurate, predictive modeling of arbitrary geometry integrated resistor structures manufactured in a variety of technologies, including those of both multichip modules (MCM's) and integrated circuits (IC's). The technique is based upon generating test structures in the process of interest, performing measurements, and extracting the behavior of a few key well identified building blocks. These building blocks can then be used for generating circuit models of other any structure created by valid combinations of those building blocks, which can then be simulated in a standard circuit simulator to predict behavior. The procedure has been experimentally verified, and shows good agreement with actual measurements up to 5-10 GHz. In addition, the model validity has been tested in several circuits by comparing the model predicted results against results obtained using the HP MDS simulator which uses measured parameters directly, with very good results. Since lumped element circuits are generated by this method, structure prediction speed is determined by circuit size and simulator small signal analysis time. The method is versatile and is well suited for circuit design applications.

    Keywords:
    Multichip modules;Semiconductor device models;Resistors;Thin film devices;Computer simulation;Serpentine;


Duke University * Pratt * Reload * Login