Computer-aided design and analysis of circuits and semiconductor – Nanotechnology related integrated circuit design
Reexamination Certificate
2001-04-23
2004-03-23
Smith, Matthew (Department: 2825)
Computer-aided design and analysis of circuits and semiconductor
Nanotechnology related integrated circuit design
C703S002000, C703S016000, C324S762010, C702S059000, C702S109000, C702S118000, C702S120000
Reexamination Certificate
active
06711723
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a method for modeling semiconductor devices and more particularly to a method for modeling semiconductor devices, such as field effect transistors (FET) and high electron mobility transistors (HEMT) which utilizes a semi-physical device model along with a data fitting model to form a relatively accurate physical device model for the use in large-signal microwave circuit computer aided design (CAD) tools.
2. Description of the Prior Art
HEMT technology provides unparalleled, high-performance characteristics at high frequencies (microwave to millimeter wave). As such, HEMTs are used in various RF applications. In order to accurately forecast the performance of such devices it is necessary accurately model the effect of the components physical structure on its high frequency, nonlinear and large signal characteristics. Thus, it is necessary to know how physical changes to the device will effect device performance in order to determine what process changes may be acceptable to improve RF yield product and which may be unacceptable which decrease RF product yield.
Large signal microwave circuit computer aided design (CAD) is used to predict the performance of a modeled circuit at high microwave input power levels. The bulk of large signal microwave circuit CAD is based upon a harmonic balance method, which is a frequency domain analysis. This method demands that elements within the circuit satisfy self-consistent conditions, such as charge conservation for all steps within the method's algorithm.
Currently, there are two predominant methods for modeling large signal characteristics of HEMT devices: table-based device modeling; and data-fitting non-linear device modeling.
Table-based device modeling utilizes measured, bias-dependent small signal characteristics for a device along with software-based algorithms to represent a full large signal model for the device. Because the measured data usually spans almost the entire range of the possible device biases, a large table of data must be created. As such, this method is known as a table based or database model. Such table-based models are disclosed in: “A Simplified Broad-Band Large Signal Non-Quasi-Static Table-Based FET Model,” by M. Fernandez-Barciela, et al.,
IEEE Trans., Microwave Theory Tech
, Vol. 48, No. 3, March 2000. The algorithms that make up this method are designed to produce modeling expressions that satisfy the charge conservative constraint required in a harmonic balance simulation. Thus, table-based method are known to perform well in large signal microwave circuit CAD.
Although the table-based method has proven to be accurate and versatile for large signal microwave circuit CAD, it has several disadvantages. For one, the method requires the use of special, software-based algorithms that are only available through some commercial software packages. Some of these packages include the microwave design system (MDS), advanced-design system (ADS) and ICCAP by Agilent Technologies. Secondly, this method is not based on the device physics and thus does not describe the device's performance in terms of any of its physical characteristics. Thus, the method is not able to predict device performance based upon physical data. Microwave circuit designers and device technologists have developed empirical data-fitting analytical expressions that can both fit measured DC and small signal characteristics of semiconductor devices and fit measured DC and small signal characteristics of semiconductor devices which satisfy the charge conservation law. Thus, the so-called non-linear data-fitting device models perform robustly in harmonic balance based large signal simulations and are thus ideal for use for large signal microwave circuit CAD. However, because these models are completely empirical, they are not able to relate real, physical characteristics with the predicted device performance. Examples of data-fitting non-linear models for HEMT devices are disclosed in: “A New Empirical Non-Linear Model for HEMT and MESFET Devices,” by I. Angelov,
IEEE Trans. Microwave Tech.
, Vol. 40, No. 12, December 1992. Thus, there is a need to provide a large signal microwave circuit CAD model that is able to relate physical device characteristics to its predicted performance.
SUMMARY OF THE INVENTION
Briefly, the present invention relates to a hybrid modeling method that forms models from a semi-physical device model along with an accurate data-fitting model in order to implement a relatively accurate physical device model as a model that is usable in large signal microwave circuit computer-aided design (CAD) tool. The semi-physical device model enables accurate representation of known physical device characteristics and measured bias-dependent characteristics. This model is used to accurately simulate the effect of process variation and environmental changes on bias-dependent characteristics. The data-fitting model is used to model these characteristics with relatively good fidelity. The expressions of the data-fitting model are constructed to be charge conservative. As such, the resultant model is physically representative of the device is computationally robust within the harmonic balance algorithms employed by known large signal microwave circuit CAD tools.
REFERENCES:
patent: 5047971 (1991-09-01), Horwitz
patent: 5051373 (1991-09-01), Yamada et al.
patent: 5418974 (1995-05-01), Craft et al.
patent: 5446676 (1995-08-01), Huang et al.
patent: 5452225 (1995-09-01), Hammer
patent: 5467291 (1995-11-01), Fan et al.
patent: 5535146 (1996-07-01), Huang et al.
patent: 5553008 (1996-09-01), Huang et al.
patent: 5568395 (1996-10-01), Huang
patent: 5648920 (1997-07-01), Duvvury et al.
patent: 5706206 (1998-01-01), Hammer et al.
patent: 5790415 (1998-08-01), Pullela et al.
patent: 5799172 (1998-08-01), Gullapalli et al.
patent: 5821766 (1998-10-01), Kim et al.
patent: 5838947 (1998-11-01), Sarin
patent: 5850355 (1998-12-01), Molnar
patent: 5942909 (1999-08-01), Zhu et al.
patent: 6266629 (2001-07-01), Mallavarpu et al.
patent: 6295630 (2001-09-01), Tamegaya
patent: 6381564 (2002-04-01), Davis et al.
patent: 6408425 (2002-06-01), Mizutani
patent: 6425111 (2002-07-01), del Mar Hershenson
patent: 6456947 (2002-09-01), Adamiak et al.
patent: 6472233 (2002-10-01), Ahmed et al.
patent: 6571182 (2003-05-01), Adamiak et al.
patent: 2002/0042698 (2002-04-01), Meuris et al.
Daniels et al., “A universal large/small signal 3-terminal FET model using a nonquasistatic charge-based approach”, IEEE Transactions on Electron Devices, vol. 40, No. 10, Oct. 1993, pp. 1723-1729.*
Brown et al., “SPICE model of the resonant-tunnelling diode”, Electronics Letters, vol. 32, No. 10, May 9, 1996, pp. 938-940.*
Allam et al., “Large signal model for analysis and design of HEMT gate mixer”, IEEE Microwave and Guided Wave Letters, vol. 4, No. 12, Dec. 1994, pp. 405-407.*
Angelov et al. (“An empirical-table based FET model”, 1999 IEEE MTT-S International Microwave Symposium Digest, vol. 2, Jun. 13, 1999, pp. 525-528).*
Shirakawa et al. (“A New Empirical Large-Signal HEMT Model”, IEEE Transactions on Microwave Theory and Techniques, vol. 44, No. 4, Apr. 1996, pp. 622-624).*
Tsai et al. (“Forecasting method for HEMT MMIC large-signal RF yield”, 22nd Annual GaAs IC Symposium, Nov. 5, 2000, pp. 125-128).*
Hirose et al. (“A Possible Scaling Limit for Enhancement-Mode GaAs MESFET's in DCFL Circuits”, IEEE Transactions on Electron Devices, vol. 39, No. 12, Dec. 1992, pp. 2681-2685).*
Kola et al., “An Analytical Expression for Fermi Level Versus Sheet Carrier Concentration for HEMT Modeling”, IEEE Electron Device Letters, vol. 9, No. 3, Mar. 1988, pp. 136-138.*
Chen et al., “A New Empiral I-V Model for HEMT Devices”, IEEE Microwave and Guided Letters, vol. 8, No. 10, Oct. 1998, pp. 342-344.*
Rizzoli et al., “General Noise Analysis of Nonlinear Microwave Circuits by the Piecewise Harmonic-Balance Technique”, IEEE Transactions on Microwave Theory and Techniques, vol. 42, No. 5, May 1994, pp.
Chen Yaochung
Tsai Roger S.
Kik Phallaka
Northrop Grumman Corporation
Paniaguas John S.
Smith Matthew
LandOfFree
Hybrid semi-physical and data fitting HEMT modeling approach... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Hybrid semi-physical and data fitting HEMT modeling approach..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Hybrid semi-physical and data fitting HEMT modeling approach... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3291425