中文摘要：

针对传统多通道盲源分离技术要求接收传感器数目不少于源信号数目,限制其在单通道的通信场景中使用的问题,提出一种小波包分解结合独立分量分析的单通道盲源分离跳频通信抗干扰方法。该方法利用小波包分解与重构技术,结合通信双方共享的跳频图案等先验信息,得到一路虚拟观测信号,使单路接收信号增维成伪MIMO矩阵,进而用独立分量分析算法实现扰信分离。计算机仿真结果表明,本文所提方法能够明显提高强阻塞干扰下跳频通信的通信性能,为单通道盲源分离通信抗干扰技术的实际应用提供了有益参考。

英文摘要：

The traditional blind source separation （BSS） requires that the number of sensors should not be less than that of the sources. However, this requirement could not be met when there is only one channel in communication senario. In light of this, a novel single-channel BSS method for anti-jamming FH com- munication combining wavelet packet decomposition and independent component analysis （ICA） is pro- posed. Taking advantage of wavelet packet decomposition and reconstruction technology and in combination of the prior-knowledge of FH pattern shared by both sides, this method acquires the single-channel virtual obervation signal, then expands the single-channel received signal to a pseudo-MIM0 matrix, thus the separation of interference and signal can be done via ICA algorithm. Simulation results indicate that the proposed method could obviously improve the communictaion performance of FH communication under strong noise jamming, and provide a valuable reference for the application of single-channel BSS in com- munication anti-jamming technology.