针对传统独立成分分析算法存在的不足,在简要介绍独立成分分析的基本原理和相关算法的基础上,提出一种结合负熵与信噪比的独立成分分析法,推导了算法的关键公式,给出了实现算法,并进行了计算机仿真实验,分别使用传统算法和改进算法对模拟产生的合成数据进行分离,通过对实验结果进行的计算分析表明了所提出的改进算法比基于负熵的传统算法具有更佳的信号分离能力,能更好地从混合信号中估计出源信号。 Aiming at the shortcomings of the traditional independent component analysis algorithms, this paper briefly introduces the basic principles of independent component analysis and related algorithms, and proposes an independent component analysis method that combines negative entropy and signal-to-noise ratio. The key formulas are deduced. The algorithm is implemented and the computer simulation experiments are carried out. The traditional algorithm and the improved algorithm are respectively used to separate the synthetic data generated by the simulation. The calculation and analysis of the experimental results show that the proposed improved algorithm is better than the traditional algorithm based on negative entropy With better signal separation capability, it is better to estimate the source signal from the mixed signal.