Diese Seite wird nicht mehr aktualisiert. Bitte besuchen Sie unsere neue Webpräsenz.

This page is not updated any longer. Please visit our new website.

Marie Curie Industry Host Fellowships

O-MOORE-NICE!

Operational MOdel Order REduction for Nanoscale IC Electronics

Call Identifier: FP6-2005-Mobility-3

Simulation plays a major role in computer aided design of integrated circuits (ICs). Mathematical models describe the dynamical processes and interactions of electrical devices. Verification of a circuit's behaviour by means of solving these model equations in time and frequency domain is a mandatory task in the design process. The structures' sizes are decreasing, the packing density gets higher and so do the driving frequencies. This requires to use refined models and take into account secondary, parasitic effects. The very high dimensional problems that emerge in this way may be solvable with the help of computer algebra in an unreasonable amount of time only. Clearly, this conflicts with the short time-to-market demands in industry.
Model-Order-Reduction presents a way out of this dilemma. Redundancies are resolved, less relevant quantities are replaced by the most significant ones. In this way, the problem's complexity is reduced, keeping the main characteristics. Solving lower dimensional problems one can get statements on the circuit's performance more quickly. Knowledge and experience in this field of computational science is scattered across european universities and companies.

In this three years' transfer-of-knowledge (ToK) project, algorithmic knowledge from university is combined with application knowledge held at industry. In this way, theoretical approaches can be verified with real life problems from semiconductor industry. The industrial partner benefits from new ideas from academia. Furthermore, activities held at the universities provide an insight for new researchers to the tasks one is confronted with in the industry.

Four partners from industry and university work on different approaches for model-order-reduction, always trying to find synergies.

• Technical University of Chemnitz

  Linear and Nonlinear Model-Order-Reduction (MOR)

  Here, the focus is on the "classical" approaches to model order reduction, i.e., complexity reduction using e.g., balanced truncation, Krylov-techniques or POD approaches. For semiconductor applications it becomes essential to generalise current techniques to differential-algebraic equations, include design parameters, preserve stability, realise reduced systems in terms of netlists again and include nonlinearity. For the latter especially approaches based on piecewise linearisations along solution trajectories (TPWL) and interpolation techniques for nonlinear parts are investigated.
  
  

  Project director:

  Prof. Dr. Peter Benner

  Researcher:

  Dr. Michael Striebel
  
  

• University of Antwerpen

  Response Surface Modelling (RSM)

  Due to the decreasing structure sizes fluctuations in the manufacturing process, e.g., variations of the layer thickness of the silicon waver and in the photolithography, can have a large influence on the performance. This necessitates the use of robust design techniques. Here response surface modelling (RSM) can be employed to reflect the dependency of a circuit's behaviour from a large amount of parameters. RSM is a data driven approach where statements on a circuit quantity for a specific set of parameters are derived from a wisely chosen collection of data. Vector fitting, e.g., is a technique from response surface modelling.
  Topics like parameterization, parameter screening, nonlinearity, and combination of approaches from optimization and statistics, are the ingredients to be studied.
  
  

  Project director:

  Prof. Dr. Tom Dhaene
  

  Researcher:

  Dr. Luciano De Tommasi
  

• Technical University of Eindhoven

  Behavioural Model Order Reduction (BMOR)

  A top-down, system design approach is needed in which detailed block descriptions are combined with behavioural models, to early study effects on integration and functionality. Behavioural MOR (BMOR), including sensitivities with relation to design, process and physical parameterizations, requires new research to timely obtain behavioural models.
  
  

  Project director:

  Prof. Dr. Wil Schilders
  

  Researcher:

  Dr. Davit Harutyunyan
  

• NXP Semiconductors, Eindhoven

  Industrial partner; Work on MOR, RSM, BMOR

  The industrial partner serves as a common platform for the researchers working on the different disciplines of model order reduction. During the first two years they stay here to integrate, test and enhance approaches, keeping in view the needs in semiconductor applications, e.g., hierarchical model-structures, different simulation modes (like transient-, ac-, pss-analysis). Besides preserving a tight connection to the partners from academia and exploiting synergies the focus of research done here directly lies on linear and nonlinear MOR.
  
  

  Project director:

  Dr. E. Jan W. ter Maten

  E-mail:jan.ter.maten(at)nxp.com

  Researcher:

  Dr. Joost Rommes

  E-mail: joost.rommes(at)nxp.com

---

Project Manager:

Feb. 01, 2007 - Dec. 14, 2008, Ir. Marcel F. Sevat, NXP Semiconductors, Eindhoven

E-mail: marcel.sevat(at)nxp.com

since Dec. 15, 2008, Dr. E. Jan W. ter Maten, NXP Semiconductors, Eindhoven

E-mail: Jan.ter.Maten(at)nxp.com

---

Publications

2010

O-MOORE-NICE! New methodologies and algorithms for design and simulation of analog integrated circuits. Rommes, J.; Harutyunyan, D.; Striebel, M.; De Tommasi, L.; ter Maten, E.J.W.; Benner, P.; Dhaene, T.; Schilders, W.H.A.; Sevat, M. ECMI Newsletter, Number 47, March 2010, pp. 17-20.

Computing Rightmost Eigenvalues for Small-Signal Stability Assessment of Large-Scale Power Systems. Rommes, J.; Martins, N.; Freitas, F. to appear in: IEEE Transactions on Power Systems. 2010.

Efficient methods for large resistor networks. Rommes, J.; Schilders, W.H.A. IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, Vol. 29, Issue 1, January 2010, pp. 28-39.

Model Order Reduction for nonlinear network problems. Striebel, M.; ter Maten, E.J.W. to appear in: COMSON Handbook, 2010.

Surrogate Modeling of Low Noise Amplifiers based on Transistor Level Simulations. De Tommasi, L.; Gorissen, D.; Croon, J.; Dhaene, T. in: Janne Roos and Luis R.J. Costa (Eds.) Scientific Computing in Electrical Engineering SCEE 2008, Mathematics in Industry, Vol. 14, Springer-Verlag, Berlin/Heidelberg, pp. 225-232, 2010.

Model Order Reduction for Nonlinear IC Models with POD; Verhoeven, A., Striebel, M.; ter Maten, E.J.W. in: Janne Roos and Luis R.J. Costa (Eds.) Scientific Computing in Electrical Engineering SCEE 2008, Mathematics in Industry, Vol. 14, Springer-Verlag, Berlin/Heidelberg, pp. 571-578, 2010.

Simulation of large interconnect structures using ILU-type preconditioner. Harutyunyan, D.; Schoenmaker, W.; Schilders, W. H. A. in: Janne Roos and Luis R.J. Costa (Eds.) Scientific Computing in Electrical Engineering SCEE 2008, Mathematics in Industry, Vol. 14, Springer-Verlag, Berlin/Heidelberg, pp. 395-402, 2010.

Reduction of Large Resistor Networks. Rommes, J.; Lenaers, P.; Schilders, W.H.A. in: Janne Roos and Luis R.J. Costa (Eds.) Scientific Computing in Electrical Engineering SCEE 2008, Mathematics in Industry, Vol. 14, Springer-Verlag, Berlin/Heidelberg, pp. 555-562, 2010.

A New Approach to Passivity Preserving Model Reduction: the Dominant Spectral Zero Method. Ionutiu, R.; Rommes, J.; Antoulas, A.C. in: Janne Roos and Luis R.J. Costa (Eds.) Scientific Computing in Electrical Engineering SCEE 2008, Mathematics in Industry, Vol. 14, Springer-Verlag, Berlin/Heidelberg, pp. 491-498, 2010.

Advances in Balancing-Related Model Reduction for Circuit Simulation. Benner, P. in: Janne Roos and Luis R.J. Costa (Eds.) Scientific Computing in Electrical Engineering SCEE 2008, Mathematics in Industry, Vol. 14, Springer-Verlag, Berlin/Heidelberg. pp.469-482, 2010.

2009

Model order reduction for nonlinear IC models. Verhoeven, A.; ter Maten, E.J.W.; Striebel, M.; Mattheij, R. in: A. Korytowski, K. Malanowski, W. Mitkowski, M. Szymkat (Eds.): System Modeling and Optimization, IFIP AICT 312, Springer-Verlag, pp. 476-491, 2009.

Sequential Modeling of a Low Noise Amplifier with Neural Networks and Active Learning. Gorissen, D.; De Tommasi, L.; Crombecq, K.; Dhaene, T. Neural Computing & Applications (Springer), Vol. 18, Nr. 5, pp. 485-494, June 2009.

Transfer Function Identification from Phase Response Data. De Tommasi, L.; Deschrijver, D.; Dhaene, T. to appear in: AEÜ International Journal of Electronics and Communications, 2009.

Robust Transfer Function Identification via an Enhanced Magnitude Vector Fitting Algorithm. De Tommasi, L.;, B. Gustavsen, B.; Dhaene, T. to appear in: IET Control Theory and Applications, 2009

A Multirate ROW-Scheme for Index-1 Network Equations. Striebel, M.; Bartel, A.; Günther, M. in Appl. Numer. Math. 59 (2009), 3-4, pp. 800-814

Forward and reverse modeling of low noise amplifiers based on circuit simulations. De Tommasi, L.; Rommes, J.; Beelen, T.; Sevat, M.; Croon, J.A.; Dhaene, T. to appear in: In P. Benner, M. Hinze, E.J.W. ter Maten (eds.), Model Reduction for Circuit Simulation, Lecture Notes in Electrical Engineering, Springer-Verlag, Berlin/Heidelberg.

Simulation of coupled oscillators using nonlinear phase macromodels and model order reduction. Harutyunyan, D.; Rommes, J. to appear in: In P. Benner, M. Hinze, E.J.W. ter Maten (eds.), Model Reduction for Circuit Simulation, Lecture Notes in Electrical Engineering, Springer-Verlag, Berlin/Heidelberg.

Model Reduction of Periodic Descriptor Systems Using Balanced Truncation. Benner, P.; Hossain, M.-S.; Stykel, T. to appear in: In P. Benner, M. Hinze, E.J.W. ter Maten (eds.), Model Reduction for Circuit Simulation, Lecture Notes in Electrical Engineering, Springer-Verlag, Berlin/Heidelberg.

Simulation of mutually coupled oscillators using nonlinear phase macromodels. Harutyunyan, D.; Rommes, J.; ter Maten, E.J.W.; Schilders, W.H.A. IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems (TCAD), 2009, Volume 28, Issue: 10, pp. 1456-1466. also: TUE-CASA Report 08-42

Exploiting structure in large-scale electrical circuit and power system problems. Rommes, J.; Martins, N. Linear Algebra and its Applications, Vol. 431, Issue 3-4, 2009, pp. 318-333.

Model order reduction for semi-explicit systems of differential algebraic systems.. Mohaghegh, K.; Pulch, R.; Striebel, M.; ter Maten, E.J.W. in: I. Troch, F. Breitenecker (Eds): Proceedings MATHMOD 09 Vienna - Full Papers CD Volume, ISBN 978-3-901608-35-3, pp. 1256-1265, 2009.

Preconditioning techniques in linear model order reduction. Schilders, W.H.A.; Rommes, J. Proceedings MATHMOD 09 Vienna, pp. 1223-1231, 2009

Reduction of large resistor networks. Rommes, J.; Lenaers, P.; Schilders, W.H.A. CASA-report 09-38, Technische Universiteit Eindhoven, 2009.

Model order reduction for multi-terminal circuits. Ionutiu, R.; Rommes, J. CASA-report 09-29, Technische Universiteit Eindhoven, 2009

A framework for synthesis of reduced order models. Ionutiu, R.; Rommes, J. CASA-report 09-28, Technische Universiteit Eindhoven, 2009.

Simulation of coupled oscillators using nonlinear phase macromodels and model order reduction. Harutyunyan, D.; Rommes, J. CASA-report 09-08, Technische Universiteit Eindhoven, 2009.

2008

Accurate Macromodeling Based on Tabulated Magnitude Frequency Response. De Tommasi, L.;, B. Gustavsen, B.; Dhaene, T. in proceedings: 12th IEEE Workshop on Signal Propagation on Interconnects (SPI 2008), Avignon (France), May 2008.

Single-Input-Single-Output Passive Macromodeling via Positive Fractions Vector Fitting. De Tommasi, L.; Deschrijver, D.; Dhaene, T. to appear in: proceedings of 12th IEEE Workshop on Signal Propagation on Interconnects (SPI 2008), Avignon (France), May 2008.

RF circuit block modeling via Kriging surrogates. Gorissen, D.; De Tommasi, L.; Hendrickx, W.; Croon, J.; Dhaene, T. in proceedings: 17th International Conference on Microwaves, Radar and Wireless Communications (IEEE MIKON-2008), Kraków (Poland), pp. 688-691, May 2008

Automatic Model Type Selection with Heterogeneous Evolution: An application to RF circuit block modeling. Gorissen, D.; De Tommasi, L.; Croon, J.; Dhaene, T. to proceedings:IEEE World Congress on Computational Intelligence (WCCI 2008), Hong Kong (China), pp. 989-996, June 2008.

Surrogate modeling of RF circuit blocks. De Tommasi, L.; Gorissen, D.; Croon, J.; Dhaene, T. to appear in proceedings: ECMI 2008, Springer-Verlag.

A Novel Sequential Design Strategy for Global Surrogate Modeling. Crombecq, K.; Gorissen, D.; De Tommasi, L.; Dhaene, T. to appear in proceedings: 41th Conference on Winter Simulation, Austin, Texas, USA.

Model order reduction of nonlinear systems: status, open issues, and applications. Striebel, M.; Rommes, J. in Chemnitz Scientific Computing Preprints CSC 08-07

Hierachical Mixed Multirating in Circuit Simulation. Striebel, M.; Bartel, A.; Günther, M. to appear in: Proceeding of the ECMI 2008, London

Model order reduction of nonlinear systems in circuit simulation: status and applications. Striebel, M.; Rommes, J. to appear in: In P. Benner, M. Hinze, E.J.W. ter Maten (eds.), Model Reduction for Circuit Simulation, Lecture Notes in Electrical Engineering, Springer-Verlag, Berlin/Heidelberg.

Projection-Based Model Reduction for LTV Descriptor System Using Multipoint Krylov-Subspace Projectors. Hossain, M.S.; Benner, P. Proceedings in Applied Mathematics and Mechanics, Vol. 8, No. 1, pp. 10081-10084, 2008. DOI: 10.1002/pamm.200810081

Passivity-Preserving Model Reduction Using Dominant Spectral-Zero Interpolation. Ionutiu, R.; Rommes, J.; Antoulas, A. C. IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, Vol. 27, Issue 12, December 2008, pp. 2250-2263.

Gramian-Based Reduction Method Applied to Large Sparse Power System Descriptor Models. Freitas, F.; Rommes, J.; Martins, N. IEEE Transactions on Power Systems, Vol. 23, Issue 3, August 2008, pp. 1258-1270.