Enhanced ICA-R Algorithm Based Steady Mixed Signals Separation

To solve the problem of steady mixed signal separation, we propose an enhanced reference independent component analysis (ICA-R) method by combining the negative entropy contrast function and the distance measurement function of the target signal. The new algorithm transforms the constrained optimization problem into an unconstrained optimization problem. The convergence of the enhanced ICA-R method is proved theoretically. Simulation results show that the algorithm proposed in this paper improves the robustness of target signal extraction.