Constant Modulus Blind Equalization for Time-Varying MIMO-FIR Channels with Pulse Estimation

The constant modulus (CM) property of source signal can be mostly utilized to equalize the singleinput multiple-output and finite impulse response (FIR) channels. In existing, the channels are blindly equalized and the equalization methods are based on higher order statistics that estimates all non-zero channel pulsation under time-invariance assumption. However, the assumption cannot be satisfied time varying multiple input multiple output (MIMO) applications ex. Mobile communication. In this paper, the proposed algorithm which extends the CM criterion to blind equalization using complex exponential basic expansion model (CEBEM) and the channel is assumed as time varying MIMO-FIR. The method only employ the Second order statistics (SOS) and finally, it estimates only one pulsation. In this way, the system increases the SNR of the transmitted symbols and produces most beneficial result in time-varying channels. The fast convergence is also achieved through zero forcing equalization.