Algorithm of Adaptive Online Blind Signal Separation Based on Matrix Separation

For the studied the over-determined blind source separation algorithm and its application in radar signal sorting. The reason that natural gradient algorithm of over-determined BSS can not stably converge eventually is analyzed. Aimed at the problem that the exiting methods are analyzed and researched. On-line BSS algorithm with adaptive step length based on separating matrix is presented to implement the optimum combination between the convergence speed and the steady-state error. At the same time, the algorithm can achieve a better separating result when the signal is randomly reduced or increased. The simulation result verifies the convergence stability and the separating effectivity of the two improved algorithms.