Application of Different Basis and Neural Network Turbo Decoding Algorithm in Multicarrier Modulation System over Time-Variant Channels

In this paper filter bank generated by different orthogonal Daubechies wavelets and bi-orthogonal wavelets are tested. Due to the high spectral containment of wavelet filters, the wavelet packet Multicarrier moduclation (WP-MCM) system performs better than Fourier based MCM system over time-variant channels. Turbo code adopting neural network (NN) decoding algorithm is applied in WP-MCM system, the BER performance improves greatly over time-variant channels with few iterations in the regions of low signal to noise ratio and is very close to that adopting the maximum a posteriori (MAP) decoding algorithm.

[1]  G.L. Stuber,et al.  Interchannel interference analysis of OFDM in a mobile environment , 1995, 1995 IEEE 45th Vehicular Technology Conference. Countdown to the Wireless Twenty-First Century.

[2]  Ingrid Daubechies,et al.  Ten Lectures on Wavelets , 1992 .

[3]  Harry C. S. Rughooputh,et al.  Neural network decoding of turbo codes , 1999, IJCNN'99. International Joint Conference on Neural Networks. Proceedings (Cat. No.99CH36339).

[4]  H. Nikookar,et al.  Wavelet based OFDM for wireless channels , 2001, IEEE VTS 53rd Vehicular Technology Conference, Spring 2001. Proceedings (Cat. No.01CH37202).

[5]  A. Glavieux,et al.  Near Shannon limit error-correcting coding and decoding: Turbo-codes. 1 , 1993, Proceedings of ICC '93 - IEEE International Conference on Communications.