A Software Framework for Automated Behavioral Modeling of Electronic Devices

被引:4
作者
Gorissen, Dirk [1 ]
Deschrijver, Dirk
Dhaene, Tom [1 ]
De Zutter, Daniel [1 ]
机构
[1] Univ Ghent, IBBT, Dept Informat Technol INTEC, B-9050 Ghent, Belgium
关键词
Behavioral science; Modeling; Design methodology; Electronic components; Aerodynamics; Mechanical engineering; OPTIMIZATION; ALGORITHM; DESIGN; APPROXIMATION; STABILITY; PASSIVITY; STRATEGY;
D O I
10.1109/MMM.2012.2205836
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
A unified and automated modeling framework is presented that can assist an electronic design automation (EDA) domain expert in generating accurate behavioral models. The framework is a SUMO Toolbox, a flexible tool that integrates different modeling approaches and implements an adaptive behavioral model construction algorithm based on the flowchart. On the modeling side, the model generation process takes full advantage of multicore CPUs if the MATLAB Parallel Computing Toolbox is available, and even of a complete cluster or grid if the MATLAB Distributed Scheduler is available. The performance of the framework is demonstrated with the help of two EM device modeling problems, a passive component and an active device. Results show that ANN model generation code is able to reduce the true error evaluated over a dense reference grid quite effectively while minimizing the estimated error. Obtaining the required circuit design parameters can be done through an approximation of the circuit performance figures based on one or more behavioral models.
引用
收藏
页码:102 / 118
页数:9
相关论文
共 51 条
  • [1] Dynamic System Equivalents: A Survey of Available Techniques
    Annakkage, U. D.
    Nair, N. K. C.
    Liang, Yuefeng
    Gole, A. M.
    Dinavahi, V.
    Gustavsen, Bjorn
    Noda, Taku
    Ghasemi, Hassan
    Monti, A.
    Matar, Mah
    Iravani, R.
    Martinez, J. A.
    [J]. IEEE TRANSACTIONS ON POWER DELIVERY, 2012, 27 (01) : 411 - 420
  • [2] [Anonymous], THESIS POLYTECH U TU
  • [3] [Anonymous], NEURAL NETWORKS RF M
  • [4] [Anonymous], 2004, THESIS GEORGIA I TEC
  • [5] [Anonymous], HDB RES GRID TECHNOL
  • [6] [Anonymous], DASE DESIGN ANAL SIM
  • [7] [Anonymous], P INT JOINT C NEUR N, DOI DOI 10.1109/ICNN.1997.614194
  • [8] [Anonymous], P 7 NIPS C ADV NEUR
  • [9] [Anonymous], LEARNING DATA CONCEP
  • [10] Broadband rational macromodeling based on the adaptive frequency sampling algorithm and the partial element equivalent circuit method
    Antonini, Giulio
    Deschrijver, Dirk
    Dhaene, Tom
    [J]. IEEE TRANSACTIONS ON ELECTROMAGNETIC COMPATIBILITY, 2008, 50 (01) : 128 - 137