This paper presents Adaptive Blind Identification of SIMO Channels. In this, the problem of identifying a linear as well as nonlinear system without a training signal, the so-called blind method is addressed using three adaptive multi-channel approaches i.e. Least mean square (MCLMS), Newton (MCL) and Recursive least squares (MCRLS) algorithms. The proposed algorithms are used for blind system identification using a magnetic recording model. In MCLMS and MCN algorithms, error is calculated based on cross relation between system output and system parameters where as RLS is based on the one-step forward linear prediction (LP) technique. It is observed that NRMSE gets reduced with increased SNR in all algorithms but the MCRLS algorithm dramatically accelerates the convergence. Convergence of the MCRLS is better than MCLMS algorithm. Convergence of the MCN is faster than both MCRLS and MCLMS. The new technique used for nonlinear system identification is working satisfactorily even at low signal to noise ratio (SNR) values. It is observed that the proposed algorithms converge in the mean to the desired channel impulse responses for an identifiable system.
