Signal Feature Extraction And Matching In The Blind Scout

A new system of communication counterwork and advanced technology must be bring forward to satisfy the requirement of modernization war. In this paper, a unique model of communication counterwork based on blind scout and blind interference was constructed firstly, and then some key components in blind system such as feature extraction and matching was deeply researched, subsequently, a new method of signal identification was proposed in this paper.The new method avoids distinguishing the signal modulation mode, but selects the instantaneous parameter of the signal as basic research object. The received signal was firstly disassemble into several IMF components by a new time frequency method named HHT to guarantee the instantaneous parameter has the enough physical meaning, and then an instantaneous character matrix was constructed using IMFs to reduce the number of the instantaneous parameter, the result was stored in signal feature library as instantaneous feature vector. At last, mode identification or neural network was used for determining whether the received signal is the same as one of reference signal in the feature library. The finally results of experiment indicate that signal identification we proposed in this paper is not only fit for known signals but also has favorable performance for unknown signals.