Least Squares Support Vector Machines for Channel Prediction in the MIMO System - Developing Concepts in Applied Intelligence - page 44

or after some algebraic manipulations

σ e = E | u 1 UDV H ˆ

2

(19)
The RBF kernel with the width parameters: σ =0 . 5 and the regularizotion
parameter γ = 500 were selected, respectively.
The obtained values of the maximum singular value of the error matrix and
the minimum singular value of the error matrix are given in Fig. 1 and Fig. 2,
respectively.

(ˆ v 1 − σ max ) x |

5C l i n

A recurrent LS-SVM method was used in order to predict a MIMO channel. The
received SNR for an un-coded and beam-forming MIMO system was derived.
The proposed solution does not need to use the analytic mathematical model
of performance measures. However, it can provide the parameters of the system
in the varying time horizon. The experiment results show that the proposed
method achieves the best performance for the RBF kernel function. This kernel
function demands the use of scaling factors for all input parameters.

References

[1] Cortes, C., Vapnik, V.: Support Vector Networks. Machine Learning 20, 273-297
(1995)
[2] Fernandez-Getino Garcia, M.J., Rojo- Alvarez, J.L., Alonso-Atienzo, F., Martinez-
Ramon, M.: Support Vector Machines for Robust Channel Estimation in OFDM.
IEEE Signal Processing Letters 13(7), 397-400 (2006)
[3] Jindal, N., Vishwanath, S., Goldsmith, A.: On the Duality of Gaussian Multiple-
Access and Broadcast Channels. IEEE Trans. on Information Theory 50(5),
768-783 (2004)
[4] Rahman, S., Saito, M., Okada, M., Yamamoto, H.: An MC-CDMA Signal Equal-
ization and Detection Scheme Based on Support Vector Machines. In: Proc. 1st
Int. Symp. Wireless Communication Systems, pp. 11-15 (2004)
[5] 3GPP TR25.869 Tx Diversity Solutions for Multiple Antennas, v1.2.0 (August
2003)
[6] Sanchez-Fernandez, M.P., de Prado-Cumlido, M., Arenas-Garcia, J., Perez- Cruz,
F.: SVM Multiregression for Nonlinear Channel Estimation in Multiple-Input
Multiple-Output Systems. IEEE Trans. on Signal Processing 52(8), 2298-2307
(2004)
[7] Suykens, J., Lukas, L., Vandewalle, J.: Least Squares Support Vector Machine
Classifiers. Neural Processing Letters 8(3), 293-300 (1999)
[8] Suykens, J.A.K., Vandewalle, J.: Recurrent Least Squares Vector Machines. IEEE
Trans. Circuits Systems and Systems I 47(7), 11109-11114 (2000)
[9] Suykens, J.A.K., Van Gestel, T., De Brabantter, J., De Moor, B., Vandewalle,
J.: Least Squares Support Vector Machines. World Scientific Pub. Co., Singapore
(2002)
[10] Vapnik, V.: Statistical Learning Theory. John Wiley and Sons, New York City
(1998)
[11] Zheng, Y., Xiao, C.: Improved Models for the Generation of Multiple Uncorrelated
Rayleigh Fading Waveforms. IEEE Trans. Commun. Letters 6(6) (2002)