基于小波包和RobustICA的无线电混合信号分离

Separation of mixed radio signals based on wavelet packet and RobustICA

针对复杂电磁环境下无线电混合信号分离困难的问题,提出将小波包和鲁棒性独立分量分析(Robust ICA)算法应用于较低信噪比且频率接近的无线电混合信号的分离。首先用小波包分析方法对混合信号进行降噪预处理,然后采用盲源分离算法中的鲁棒性独立分量分析算法对降噪后的混合信号进行分离,通过观察分离后信号的波形和频率以及相似系数对分离结果进行定性和定量分析。所提算法与单独采用Robust ICA算法的结果对比表明:所提算法分离出的信号在波形和频率以及相似系数方面均比单独采用Robust ICA算法取得的效果好,从而证明所提算法可以较好地应用于无线电混合信号的分离。

Aiming at the difficulty of separating mixed radio signals in the complex electromagnetic environment, the wavelet packet and the robust independent component analysis （RobustICA） algorithm is proposed for the separation of mixed radio signals with lower SNR and adjacent frequency. Firstly, wavelet packet analysis is used for the noise reduction pre-processing of the mixed signals. Then, the RobustICA algorithm based on blind source separation （BSS） algorithm is used to separate the mixed signals after noise reduction. And the results are qualitatively and quantitatively analyzed by observing the waveform and frequency of the signal and the similarity coefficient. The comparison of results obtained with the proposed algorithm and the application of independent RobustICA algorithm show that the proposed algorithm has better performance than the independent RobustICA algorithm in terms of waveform, frequency and similarity coefficient. And it is proved that the proposed algorithm can be applied to the separation of mixed signals well.