基于特征提取与统计的信号类型识别

针对空间电子对抗领域的截获信号类型识别问题，提出了一种基于特征提取与统计的非合作信号类型识别方法。首先，通过对瞬时频率和复包络模值特征进行统计，实现了信号类型的粗分类，接着，基于细微特征提取和比对，实现了信号类型的精确识别。仿真结果表明，当处理器输入信噪比高于11dB时，所提方法对8类信号的正确识别概率均超过90%。本方法处理复杂度较低，易于星上工程实现。

Signal Recognition Based on Features Extraction and Statistics

Based on features extraction and statistics, this paper proposes a method to solve the signal recognition prob-lem in the field of space electronic countermeasure. Firstly, the received signal is coarsely classified according to the fea-tures of its instantaneous frequency and complex envelope module. Then, the signal is precisely identified based on the indi-vidual unique feature comparing with 8 given candidates. Simulation results show that when the input SNR （ Signal-to-Noise-Ratio） is greater than 11 dB, the recognition probility of the proposed method is more than 90% for those 8 candidate signal types. The complexity of the method is relatively low and is easy to be implemented in engineering.