Constrained optimization methods for blind equalization of multiple FIR channels

We propose a novel blind algorithm to design linear multichannel equalizers directly, by minimizing the output variance subject to appropriate constraints. The constraints are chosen to guarantee no desired signal cancellation, and their parameters are jointly optimized to maximize the signal component at the output. The resulting blind algorithm is shown to have near optimal performance at high SNR. Simulation results support our claims and indicate a superior performance when compared with subspace based channel estimation methods and multichannel linear prediction methods.