在固定步长的ICA极大似然估计自适应算法的基础上,通过一维搜索引入了步长修正方案,使新算法可在收敛速度和稳定状态时的失调误差这两个性能指标上达到最佳结合点,具有较好的时变系统跟踪能力。仿真结果证实了本文所提出的算法可以有效地提高ICA的自适应性,能够更准确地完成盲源分离。在此基础上将算法用在时变性很强的股票数据上,以验证该算法的有效性和可行性。 Based on the ICA maximum likelihood estimation adaptive algorithm with fixed step size, the step correction scheme is introduced through one-dimensional search so that the new algorithm can reach the two performance indexes of convergence speed and offset error in steady state The best combination of points, with better time-varying system tracking capabilities. The simulation results verify that the proposed algorithm can effectively improve the ICA’s self-adaptability and can more accurately complete the blind source separation. On this basis, the algorithm is applied to the stock data which is highly time-varying to verify the validity and feasibility of the algorithm.