SciELO - Scientific Electronic Library Online

 
vol.6 número3State variables monitoring using a class of nonlinear observer based estimator, applied to continuous bio-systemAlternative methods of calculation of the pseudo inverse of a non full-rank matrix índice de autoresíndice de assuntospesquisa de artigos
Home Pagelista alfabética de periódicos  

Serviços Personalizados

Journal

Artigo

Indicadores

Links relacionados

  • Não possue artigos similaresSimilares em SciELO

Compartilhar


Journal of applied research and technology

versão On-line ISSN 2448-6736versão impressa ISSN 1665-6423

J. appl. res. technol vol.6 no.3 Ciudad de México Dez. 2008

 

Evaluation algorithm for a decomposed simplicial piecewise-linear formulation

 

V.M. Jiménez-Fernández1, J. Agustín-Rodríguez2, P. Marcelo-Julián2, O. Agamennoni2

 

1 National Institute for Astrophysics, Optics and Electronics, Department of Electronics, Luis Enrique Erro No.1, Santa María Tonantzintla, P.O. 72840 Puebla, México.

2 Universidad Nacional del Sur, Departamento de Ingeniería Eléctrica y de Computadoras, Av. Alem No. 1253, P.O. 8000 Bahía Blanca, Argentina.

 

ABSTRACT

In this work an algorithm for the evaluation of N-dimensional piecewise-linear (PWL) functions is presented. The type of PWL representation which is considered is the denominated simplicial representation that is defined in a N-dimensional domain partitioned by hyperplanes and divided into simplices. The algorithm performs a local function computation into the specific simplex where the evaluation point is found. The simplicial PWL (S-PWL) interpolating equations are collected into a matrix system which adopts the form of the decomposed PWL models. The algorithm works directly with this decomposed model and determines the value of the S-PWL function simply by its values on the vertices.

Keywords: Simplicial, Piecewise-Linear, Evaluation.

 

RESUMEN

En este trabajo se presenta un algoritmo para la evaluación de funciones lineales a tramos (LAT) N-dimensionales. El tipo de representación LAT que es considerado es el denominado descripción simplicial que se define en un dominio N-dimensional particionado por hiperplanos y dividido en símplices. El algoritmo realiza un cómputo local de la función en el símplice específico donde se encuentra el punto de evaluación. Las ecuaciones de interpolación de la representación LAT simplicial (LAT-S) son colectadas en un sistema matricial que adopta la forma de los modelos LAT descompuestos. El algoritmo trabaja directamente con este modelo descompuesto y determina el valor de la función LAT-S simplemente mediante su valor en los vértices. 

 

DESCARGAR ARTÍCULO EN FORMATO PDF

 

7. REFERENCES

[1] Chua L.O. & Ying L.P., Finding all solutions of piecewise-linear circuits, International Journal of Circuit Theory and Applications, Vol. 10, July, 1982, pp. 201-229.         [ Links ]

[2] Yamamura K. & Machida A., Finding all solutions of piecewise-linear resistive circuits with high approximation accuracy, J. Circuit Syst. Comput., Vol. 15, No.3, June, 2006, pp. 389-398.         [ Links ]

[3] Leenaerts D.M.W. & Hegt J.A., Finding all solutions of piecewise-linear functions and application to circuits design, International Journal of Circuit Theory and Applications, Vol. 19, No. 2, 1991, pp. 107-123.         [ Links ]

[4] Julian P., Dogaru L. & Chua L.O., A piecewise-linear simplicial coupling cell for CNN gray-level image processing, Circuit and Systems I: Fundamental Theory and Applications, Vol. 49, No. 7 July, 2002, pp. 904-913.         [ Links ]

[5] Castro L., Figueroa J. & Agamennoni O., BIBO stability for NOE model structure using HL-CPWL functions, Proceedings of MIC.05-IASTED, 2005, Innsbruck Austria.         [ Links ]

[6] Di Federico M., Julian P., Poggi T. & Storace M., A Simplicial PWL Integrated Circuit Realization, Proceedings of IEEE International Symposium on Circuits and Systems, 2007, pp. 685-688, New Orleans, LA USA , May.         [ Links ]

[7] Sun X. & Wang S., A Special Kind of Neural Nertworks: Continuous Piecewise Linear Functions, LNCS Advances in Neural Networks, 2005, pp. 375-379.         [ Links ]

[8] Jimenez-Fernandez V.M., Rodríguez J.A., Julián P., Agamennoni O. & Di Federico M., Digital Architecture for R6 PWL Function Computation, Actas de la Escuela Argentina de Microelectrónica, Tecnología y Aplicaciones, 2007, pp. 1-6, Córdoba Arg., September.         [ Links ]

[9] Julián P., A High Level Canonical Piecewise-Linear Representation: Theory and Applications, Doctoral Thesis, Universidad Nacional del Sur, B. Blanca Argentina, 1998.         [ Links ]

[10] Julián P & Agamennoni O., High Level Canonical Piecewise-Linear Representation Using a Simplicial Partition, IEEE Transactions on Circuits and Systems-I: Fundamental Theory and Applications, Vol. 46, No. 4, April, 1999, pp. 463-480.         [ Links ]

[11] Julián P & Chua L.O., A Piecewise-Linear Simplicial Coupling Cell for CNN Gray-Level Image Processing, IEEE Transactions on Circuits and Systems-I: Fundamental Theory and Applications, Vol. 49, No. 7, July, 2002, pp. 904-913.         [ Links ]

[12] Chien M. & Kuh E., Solving Nonlinear Resistive Nerworks Using Piecewise-Linear Analysis and Simplicial Subdivision, IEEE Transactions on Circuits and Systems-I, Vol. CAS 24, No. 7, June, 1977, pp. 305-317.         [ Links ]

[13] Jiménez-Fernández V.M., Decomposed Piecewise-Linear Representation Applied to DC Analysis, Doctoral Thesis, Instituto Nacional de Astrofísica, Óptica y Electrónica INAOE, Santa María Tonantzintla Puebla, México, 2006.         [ Links ]

[14] Brzobohaty J. & Kolka Z., Decomposed Parametric Form of the State Model of a Piecewise-Linear System, IEEE International Symposium on Circuits and Systems,Vol. 6, 1994, pp. 25 - 28, June.         [ Links ]

Creative Commons License Todo o conteúdo deste periódico, exceto onde está identificado, está licenciado sob uma Licença Creative Commons