Abstract
One of the most popular meshless methods is constructed by radial kernels as basis called radial basis function method. It has a unique feature which affects significantly on accuracy and stability of approximation: existence of a free parameter known as shape parameter that can be chosen constantly or variably. Several techniques for selecting a variable shape parameter have been presented in the older works. Our study focuses on investigating the deficiency of these techniques and we introduce two new alternative strategies called hybrid shape parameter and binary shape parameter strategies based on the advantages of older studies. The proposed approaches produce the more accurate results as shown in numerical results where they are compared with random shape parameter strategy for interpolating one-dimensional and two- dimensional functions as well as in approximating the solution of Poisson equation.
Similar content being viewed by others
References
Bozzini M, Lenarduzzi L, Schaback R (2002) Adaptive interpolation by scaled multiquadrics. Adv Comput Math 16:375–387
Buhmann MD (2004) Radial basis functions: theory and implementation. University of Gissen, Cambridge University Press, Cambridge
Carlson RE, Foley TA (1991) The parameter \(R^2\) in multiquadric interpolation. Comput Math Appl 21:29–42
Cheng AH-D (2012) Multiquadric and its shape parameter-a numerical investigation of error estimate, condition number, and round-off error by arbitrary precision computation. Eng Anal Bound Elem 36:220–239
Fasshauer GE (2002) Newton iteration with multiquadratics for the solution of nonlinear PDEs. Comput Math Appl 43:423–438
Fasshauer GE (2007) Meshfree approximation methods with Matlab. In: Interdisciplinary mathematical sciences, vol 6. World Scientific Publishers, Singapore
Fasshauer GE (2011) Positive definite kernels: past, present and future. Dolomites Res Notes Approx 4:21–63
Franke R (1979) A critical comparison of some methods for interpolation of scattered data. PhD thesis, Naval Postgraduate School Monterey: California
Franke R (1982) Scattered data interpolation: tests of some methods. Math Comput 38:181–200
Gan M, Peng H, Peng X, Chen X, Inoussa G (2010) A locally linear RBF network-based state-dependent AR model for nonlinear time series modeling. Inform Sci 180:4370–4383
Golbabai A, Safdari-Vaighani A (2010) A meshless method for numerical solution of the coupled Schrödinger-KdV equations. Computing 92:225–242
Golbabai A, Rabiei H (2012a) Hybrid shape parameter strategy for the RBF approximation of vibrating systems. Int J Comput Math 89:2410–2427
Golbabai A, Rabiei H (2012b) A meshfree method based on radial basis functions for the eigenvalues of transient Stokes equations. Eng Anal Bound Elem 36:1555–1559
Golbabai A, Ahmadian D, Milev M (2012a) Radial basis functions with application to finance: American put option under jump diffusion. Math Comput Model 55:1354–1362
Golbabai A, Mohebianfar E, Rabiei H (2012b) On the role of shape parameter in approximating the Eigenvalues of fredholm integral equations: an RBF-Simpson approach. App Math Inf Sci, In Press
Hardy RL (1971) Multiquadric equations of topography and other irregular surfaces. J Geophys Res 76:1905–1915
Hon YC, Schaback R (2001) On unsymmetric collocation by radial basis function. Appl Math Comput 119:177–186
Huang CS, Lee CF, Cheng AD (2007) Error estimate, optimal shape factor, and high precision computation of multiquadric collocation method. Eng Anal Bound Elem 31:614–623
Kansa EJ (1990a) Multiquadrics—a scattered data approximation scheme with applications to computational fluid dynamics-I Surface approximations and partial derivative estimates. Comput Math Appl 19:127–145
Kansa EJ (1990b) Multiquadrics–a scattered data approximation scheme with applications to computational fluid dynamics-II solutions to parabolic, hyperbolic and elliptic partial differential equations. Comput Math Appl 19:147–161
Kansa EJ, Carlson RE (1992) Improved accuracy of multiquadric interpolation using variable shape parameters. Comput Math Appl 24:99–120
Kansa EJ, Hon YC (2000) Circumventing the ill-conditioning problem with multiquadric radial basis functions: applications to elliptic partial differential equations. Comput Math Appl 39:123–137
Kansa EJ, Aldredge RC, Ling L (2009) Numerical simulation of two-dimensional combustion using mesh-free methods. Eng Anal Bound Elem 33:940–950
Luh LT (2012) The shape parameter in the Gaussian function. Comput Math Appl 63:687–694
Madych WR (1992) Miscellaneous error-bounds for multiquadric and related interpolators. Comput Math Appl 12:121–138
Pindoriya NM, Singh SN, Sing SK (2008) An adaptive wavelet neural network-based energy price forecasting in electricity markets. IEEE Trans Power Syst 23:1423–1432
Ravi V, Kurniawan H, Thai PNK, Kumar PR (2008) Soft computing system for bank performance prediction. Appl Soft Comput 8:305–315
Rippa S (1999) An algorithm for selecting a good parameter c in radial bais function interpolation. Adv Comput Math 11:193–210
Sarra SA, Kansa EJ (2010) Multiquadric radial basis function approximation methods for the numerical solution of partial differential equations. Tech Science Press, Duluth, GA, USA
Sarra SA, Sturgill D (2009) A random variable shape parameter strategy for radial basis function approximation methods. Eng Anal Bound Elem 33:1239–1245
Schaback R (1999) Improved error bounds for scattered data interpolation by radial basis functions. Math Comput 68:201–206
Sturgill D (2009) Variable shape parameter strategies in radial basis function methods. M. Sc. Thesis, Marshall University, Huntington
Uhlir K, Skala V (2006) Radial basis function use for the restoration of damaged images. Comput Imaging Vis 32:839–844
Wendland H (2005) Scattered data approximation. Cambridge University Press, Cambridge, UK
Xiang S, Wang K, Ai Y, Sha Y, Shi H (2012) Trigonometric variable shape parameter and exponent strategy for generalized multiquadric radial basis function approximation. Appl Math Model 36:1931–1938
Yang F, Paindavoine M (2003) Implementation of an RBF neural network on embedded systems: real-time face tracking and identity verification. IEEE Trans Neural Networks 14:1162–1175
Author information
Authors and Affiliations
Corresponding author
Additional information
Communicated by Antonio José Silva Neto.
Rights and permissions
About this article
Cite this article
Golbabai, A., Mohebianfar, E. & Rabiei, H. On the new variable shape parameter strategies for radial basis functions. Comp. Appl. Math. 34, 691–704 (2015). https://doi.org/10.1007/s40314-014-0132-0
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s40314-014-0132-0
Keywords
- Meshless method
- Radial basis function
- Constant shape parameter strategies
- Variable shape parameter strategies