论文部分内容阅读
在阵列信号处理过程中，经常遇到复数信号盲分离问题。例如，卷积混合型的源信号的盲分离；声纳信号盲分离。本文提出了一类基于非线性准则的复数信号盲分离算法。将非线性函数引入学习过程，由算法自动调节学习速率。计算机仿真实验验证了算法的有效性，文中给出了验证结果。
In array signal processing, complex signal blind separation problems are often encountered. For example, blind separation of convolutional hybrid source signals; blind separation of sonar signals. In this paper, a class of complex signal blind separation algorithm based on nonlinear criterion is proposed. The nonlinear function is introduced into the learning process, and the algorithm automatically adjusts the learning rate. Computer simulation experiments verify the effectiveness of the algorithm, the paper gives the verification results.