Comparison of the channel equalizers using the inverse filter, the H2 filter and the H∞ filter

An H^∞ optimized equalizer is presented to realize an optimal causal and stable equalizer of the single-channel equalization problem in the frequency domain. The proposed algorithm achieves intersymbol interference cancellation and noise attenuation in the view point of H^∞ optimization. The communication system usually suffers some uncertainties of the channel parameter and the statistics of noise. Our algorithm is robust while the channel characteristics and the statistics of inputs and external. Noise are uncertain with a bound. Moreover, while the conventional H^∞ approach is to minimize a peak error gain over the entire range, we propose a modified H^∞ equalizer to emphasize the limited used frequency region by using a weight function. Finally, for the purpose of illustration, some examples and their simulation results are presented