An experimental investigation on a direct blind equalizer for underwater acoustic multiple FIR channel
Jianmin Liu, Yingmin Wang, Xiangrui Cao, Zhengqi Zhang · The Journal of the Acoustical Society of America · 2001
A blind equalizer removes the distortion caused by intersymbol interference (ISI) without the need for training sequences. In order to get the complete phase information of the ISI, high-order statistics have to be used to find the equalizer taps. However, L. Tong et al. have shown that most channels can be identified from the second-order cyclostationary statistics through fractionally spaced equalizers. This result has stimulated considerable research activities in fractionally spaced channel identification and equalization. In this paper a single-input multi-output (SIMO) channel blind equalizer for an underwater acoustic communication system is considered. A direct blind equalization scheme was investigated by simulations, which is a novel unsupervised adaptive equalizer and was suggested by J. Labat first for single-input single-output (SISO). An interactive matched processing (MP) is added for the SIMO case. The work verified the prediction that the linear equation form of the direct blind equalizer lends itself naturally to adaptive solutions and optimum combinations of the equalizer outputs and improved performance over existing algorithms in terms of mean-square error and probability of error in the equalized data.