A Frequency-Domain Neural Network Equalizer for OFDM
- 8 July 2004
- conference paper
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 2, 571-575 Vol.2
- https://doi.org/10.1109/glocom.2003.1258303
Abstract
OFDM is regarded as a viable solution to combat the impact of frequency selective fading; however, the channel does not have flat amplitude over the entire bandwidth, thus channel equalization is still required at the receiver. Radial basis function (RBF) neural networks have been widely considered for channel equalization, since they offer certain advantages over conventional equalizer structures. In this paper, a novel RBF channel equalizer structure, which performs Bayesian estimation, is proposed for OFDM communication systems. The proposed equalizer structure is shown to outperform existing equalizers; it can therefore be considered as a better practical alternative for OFDM channel equalization.Keywords
This publication has 9 references indexed in Scilit:
- Multi-level radial basis function network based equalisers for Rayleigh channelsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2003
- Pre- post- and balanced equalization in OFDMPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2003
- A decision directed coherent detector for OFDMPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- Channel estimation versus equalization in an OFDM WLAN systemPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- Wireless multicarrier communicationsIEEE Signal Processing Magazine, 2000
- Transmission techniques for digital terrestrial TV broadcastingIEEE Communications Magazine, 1995
- Adaptive detection of small sinusoidal signals in non-Gaussian noise using an RBF neural networkIEEE Transactions on Neural Networks, 1995
- Universal Approximation Using Radial-Basis-Function NetworksNeural Computation, 1991
- The complex LMS algorithmProceedings of the IEEE, 1975