Frequency-Hopping Network Station Sorting Based on Improved Independent Component Analvsis

Abstract

In the complex electromagnetic environment, the traditional fast independent component analysis algorithm to achieve frequency-hopping network station sorting does not consider the instantaneous frequency-hopping fast changes, complex time-frequency analysis and other characteristics, resulting in low robust performance, slow convergence, Not conducive to hardware. Therefore, based on the fast independent component analysis algorithm, the traditional nonlinear function is improved, and the Newton iteration of the fixed point algorithm is converted from second order to fifth order convergence. A new method which is suitable for the frequency-hopping signal Selected Improved Fast Independent Component Analysis Algorithm. Simulation results show that the correlation coefficient, PI value and output SNR index of the proposed algorithm are superior to the traditional fast independent component analysis algorithm, and the robustness is improved, the convergence speed is accelerated, and the computation time is reduced by more than 60%.