Blind source separation is the separation of source signals from a set of mixed signals,without or with very little information about the source signals or the mixing process.Blind source separation has become a hot spot in research of signal processing in recent years.Large amount of calculation is required for source signal recovery process in blind source separation by Kurtosis based Independent Component Analysis(ICA) algorithm,thus the conjugate gradient method is employed in this paper to optimize the ICA algorithm.The Matlab simulation results show that the improved ICA algorithm is quick in convergence speed,good in separation performance,and low in steady-state error.