基于独立分量分析的混叠跳频信号分离算法

为解决混叠跳频信号的分离问题，在深入研究独立分量分析（ICA：Independent Component Analysis）理论基础上，结合跳频通信的特点，提出了基于独立分量分析的混叠跳频信号分离算法，实现了对混叠跳频信号的盲分离。该算法将基于负熵最大化的FastlCA算法应用到混叠跳频信号分离中。通过仿真实验表明，该算法能成功地排除乘性噪声干扰，完成对混叠跳频信号的分离。虽然分离信号的幅度、相位等参数较源信号发生了变化，但并不影响后续工作。这一过程在未知任何先验参数的条件下完成，并取得了较好的分离效果，为跳频通信信号的分离工作提供了新思路。

Overlapped Frequency-Hopping Communication Signals Separation Algorithm Based on Independent Component Analysis

For the separation of overlapped frequency-hopping communication signals, a method is expounded to adopt the FastlCA （Independent Component Analysis） algorithm to separate overlapped frequency-hopping communication signals, based on researching independent component analysis theory deeply, and aiming to the characteristic of frequency-hopping Communication. It was proved that the algorithm obviated the disturbance with multiplier noise, and achieved the separation of overlapped frequency-hopping communication signals by simulating. Comparing with the source signals, the transformation of the parameter took place, such as amplitude and phase, but it didn＇t influence later work. This process obtained a good separation effect without a prior knowl- edge, so it achieved blind separation and offered a new way to separate frequency-hopping communication signals.