!i#e%-p(i)* !as*ica A./(0i*12s !(0 *1e 3.i)% Sepa0a*i() (! C(2p.e#-4a.5e% Si/)a. 2i#*50es

In this paper, we derive new xed-point algorithms for the blind separation of complex-valued mixtures of independent, possibly non-circularly-symmetric, and non-Gaussian source signals. Leveraging recent results in complex independent component analysis, we construct iterative procedures for complex signal mixtures whose evolutionary characteristics are identical to those of the real-valued FastICA algorithmwith kurtosis contrast. Our methods are fast, computationally simple, and easy to use. For extracting multiple sources, symmetric and asymmetric signal deation procedures can be employed. Simulations for both noiseless and noisy mixtures indicate that the proposed algorithms often have equal or better nite-sample performance than existing separation methods for complex-valued mixtures.