SOS-based blind equalization algorithm of FIR MIMO system in frequency domain

Blind equalization of finite impulse response multi-input multi-output system is an important task for numerous applications such as speech separation, dereverberation, communication, signal processing and control, etc. A cost function is constructed based on second order statistics of equalizer output, and a new algorithm is derived with natural gradient. The algorithm is adapted to non-stationary signals, and the learning rate is multiplied by a constraint factor, which is adjusted adaptively. Simulations indicate the ability of the algorithm to perform blind equalization under the weaker condition (the system is equalizable).