Abstract
We investigate the use of support vector machines (SVMs) to determine simpler and better fit power macromodels of functional units for high-level power estimation. The basic approach is first to obtain the power consumption of the module for a large number of points in the input signal space. Least-Squares SVMs are then used to compute the best model to fit this set of points. We have performed extensive experiments in order to determine the best parameters for the kernels.
We propose a new method for power macromodeling of functional units for high-level power estimation based on Least-Squares Support Vector Machines (LS-SVM). Our method improves the already good modeling capabilities of the basic LS-SVM method in two ways. First, a modified norm is used that is able to take into account the weight of each input for global power consumption in the computation of the kernels. Second, an iterative method is proposed where new data-points are selectively added as support-vectors to increase the generalization of the model.
The macromodels obtained compare favorably with those obtained using industry standard table models, providing not only excellent accuracy on average (close to 1% error), but more importantly, thanks to our proposed modified kernels, we were able to reduce the maximum error to values close to 10%.
Chapter PDF
Similar content being viewed by others
References
Bogliolo, A., Benini, L., De Micheli, G.: Regression-Based RTL Power Modeling. ACM Transactions on Design Automation of Electronic Systems 5(3), 337–372 (2000)
Chen, Z., Roy, K.: A power macromodeling technique based on power sensitivity. In: Proceedings of Design Automation Conference, pp. 678–683 (June 1998)
Gupta, S., Najm, F.: Power Modeling for High-level Power Estimation. IEEE Transactions on Very Large Scale Integration (VLSI) Systems 8, 18–29 (2000)
Gusmao, A., Silveira, L.M., Monteiro, J.: Parameter tuning in svm-based power macro-modeling. In: Quality of Electronic Design (ISQED 2009), pp. 135–140 (March 2009)
Hoegaerts, L., Suykens, J.A.K., Vandewalle, J., De Moor, B.: Primal space sparse kernel partial least squares regression for large scale problems. In: Proceedings of 2004 IEEE International Joint Conference on Neural Networks, vol. 1, p. 563 (July 2004)
Jochens, G., Kruse, L., Schmidt, E., Nebel, W.: A new parameterizable power macro-model for datapath components. In: Proceedings of Design, Automation and Test in Europe Conference and Exhibition 1999, pp. 29–36 (1999)
Landman, P.E., Rabaey, J.M.: Architectural power analysis: The dual bit type method. IEEE Transactions on Very Large Scale Integration (VLSI) Systems 3(2), 173–187 (1995)
Macii, E., Pedram, M., Somenzi, F.: High-level Power Modeling, Estimation, and Optimization. IEEE Transaction on Computer-Aided Design of Integrated Circuits and Systems, 1061–1079 (1998)
Marculescu, D., Marculescu, R., Pedram, M.: Information Theoretic Measures of Energy Consumption at Register Transfer Level. In: Proceedings of the International Symposium on Low Power Electronics and Design, pp. 81–86 (April 1995)
Najm, F.: Towards a High-Level Power Estimation Capability. In: Proceedings of the International Symposium on Low Power Electronics and Design, pp. 87–92 (April 1995)
Raghunathan, A., Dey, S., Jha, N.K.: High-level Macro-modeling and Estimation Techniques for Switching Activity and Power Consumption. IEEE Transactions on Very Large Scale Integration (VLSI) Systems 11(4), 538–557 (2003)
Shang, W., Zhao, S., Shen, Y.: Application of LSSVM with AGA Optimizing Parameters to Nonlinear Modeling of SRM. In: 3rd IEEE Conference on Industrial Electronics and Applications, pp. 775–780 (June 2008)
Shannon, C.: The Synthesis of Two-Terminal Switching Circuits. Bell System Technical Journal 28, 59–98 (1949)
Smits, G., Jordaan, E.: Improved SVM Regression using Mixtures of Kernels. In: Proceedings of the 2002 International Joint Conference on Neural Networks, IJCNN 2002, vol. 3, pp. 2785–2790 (2002)
Suykens, J., Van Gestel, T., De Brabanter, J., De Moor, B., Vandewalle, J.: Least Squares Support Vector Machines. World Scientific Pub., Singapore (2002)
Vapnik, V.: Statistical Learning Theory. John Wiley & Sons, Chichester (1998)
Wu, Q., Qiu, Q., Pedram, M., Ding, C.-S.: Cycle-accurate macro-models for rt-level power analysis. IEEE Transactions on Very Large Scale Integration (VLSI) Systems 6(4), 520–528 (1998)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2011 IFIP International Federation for Information Processing
About this paper
Cite this paper
Gusmão, A., Miguel Silveira, L., Monteiro, J. (2011). Power Macro-Modeling Using an Iterative LS-SVM Method. In: Becker, J., Johann, M., Reis, R. (eds) VLSI-SoC: Technologies for Systems Integration. VLSI-SoC 2009. IFIP Advances in Information and Communication Technology, vol 360. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-23120-9_7
Download citation
DOI: https://doi.org/10.1007/978-3-642-23120-9_7
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-23119-3
Online ISBN: 978-3-642-23120-9
eBook Packages: Computer ScienceComputer Science (R0)