Unconstrained optimization criteria for blind equalization of multichannel linear systems

Blind equalization and blind deconvolution have been an important interesting topic in diverse fields including data communication, image processing and geophysical data processing. Inouye and Habe (see Proc. IEEE Signal Processing Workshop on Higher-Order Statistics, p. 96, 1995) proposed a multistage maximization criterion and a single-stage maximization criterion for attaining the blind equalization of multichannel linear time-invariant systems. However their maximization criteria should be subjected to several constraints of the equations. We present an new unconstrained maximization criteria for accomplishing the blind equalization of multichannel linear time-invariant systems. Stochastic gradient algorithms are proposed for solving the unconstrained maximization problems. Simulation examples are included to examine the performance of the proposed algorithms.