Blind Separation Of Mixed Communication Signals

Current war is full of high technology means, it makes communication electronic warfare more complex. As the important component of communication electronic warfare, signal separation has the great effect increasingly. Although there are many kinds of technologies based on time-frequency could be used to separate signal, it is difficult to apply them to complex electromagnetism circumstance on field, for the sake of disadvantage - need to get the prior information to object signal. As a new means of array processing and data analysis in signal processing, blind source separation get rid off the disadvantage of signal separation technologies based on time-frequency. It has the ability to recover source signal from mixed signal without the prior information about source signal and the channel. Blind source separation could be applied to many field such as communication, iatrology, image and radar signal processing. It is currently an active-research area belonging to signal processing. In the paper, some signal separation technologies based on time-frequency and their characteristic was introduced at first. Then, generalizing blind source separation technology, involved research status in the world was summarized either. In this part, mainly discussed some blind separation algorithm, which were in commom use, and their characteristic. According to beyond works, a new algorithm for instantaneous mixed blind signal separation was put forward and extended to linear convolution mixed model. It was proved effective by simulation. At the end of paper, sumed up research on signal separation technology of modern communication electronic warfare, also viewed the prospect of its development.