A new fast nonlinear principal component analysis algorithm for blind source separation
Xin Wang, Shifeng Ou, Ying Gao, Xiaofeng Guo · 2015
The nonlinear principal component analysis (NPCA) can be applied to solve the blind source separation (BSS) problem. By combining the optimum step size with the optimum momentum factor both derived by the decrement of the cost function of NPCA algorithm, an integration fast NPCA algorithm is proposed in this paper. Simulation experiments proved that the proposed algorithm is superior to other NPCA algorithms in convergence rate and steady error.