Adaptive Convex Combination of Two RBF Networks and its Application to Nonlinear System Identification

Abstract:

This paper presents a novel nonlinear adaptive filter-based the convex combination of two radial basis function (RBF) networks (CRBF), which is trained by the stochastic gradient (SG) learning algorithm with different step sizes. The combination approach can alleviate the compromises of fast convergence and low steady-state error for RBF networks. Moreover, it is also with better tracking capability than the RBF network. Experiments with nonlinear system identification illustrate that the desired behavior and robustness.

• Periodical: Advanced Materials Research (Volumes 562-564)
• Pages: 1697-1701
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.562-564.1697
• Online since: August 2012
• Authors: Xiang Ping Zeng, Hai Quan Zhao, Wei Dong Jin
• Keywords: Neural Network (NN), Nonlinear System Identification, Radial Basis Function (RBF)

References

[1] J. Park and I. W. sandberg, Universal approximation using radial basis function networks, Neural Comput., 3(1991), pp.246-257.

DOI: 10.1162/neco.1991.3.2.246

[2] S. A. Kassam and I. Cha, Radial basis function networks in nonlinear signal processing, in Proc. IEEE 27th Annu. Asilomar Conf. Signals, Syst., Comput., 2(1993), pp.1021-1025.

DOI: 10.1109/acssc.1993.342415

[3] S. Chen, C. F. N. Cowan, and P. M. Grant, Orthogonal least squares learning algorithm for radial basis function networks, IEEE Trans. Neural Netw., 2(2)(1991), p.302–309.

DOI: 10.1109/72.80341

[4] V. Kadirkamanathan and M. Niranjan, A function estimation approach to sequential learning with neural networks, Neural Comput., 5(1993), p.954–975.

DOI: 10.1162/neco.1993.5.6.954

[5] G. B. Huang, P. Saratchandran, N. Sundararajan, An generalized growing and pruning RBF (GGAP-RBF) neural network for function approximation, IEEE Trans. Neural Networks, 16(1)(2005), pp.57-67.

DOI: 10.1109/tnn.2004.836241

[6] H. Q. Zhao, J. S. Zhang. Adaptively combined FIR and functional link neural network equalizer for nonlinear communication channel, IEEE Transactions on Neural Networks, 20(4)(2009), pp.665-674.

DOI: 10.1109/tnn.2008.2011481

[7] H. Q. Zhao, X. P. Zeng, J. S. Zhang, X. Q. Zhang, Y. G. Liu and T. Wei, Adaptive decision feedback equalizer using the combination of the FIR and FLNN, Digital Signal Processing, 21(2011), pp.679-689.

DOI: 10.1016/j.dsp.2011.05.004

[8] L. A. Azpicueta-Ruiz, M. Zeller, A. R. Figueiras-Vidal, J. Arenas-Garcia, and W. Kellermann, Adaptive combination of Volterra kernels and its application to nonlinear acoustic echo cancellation, IEEE Trans. Audio, Speech, and Language Processing, 19(1) (2011).

DOI: 10.1109/tasl.2010.2045185