基于相位差分的脉内调制信号类型识别

提出了一种工程可快速实现的常见脉内调制信号类型识别算法。首先基于相位差分重构了七种常见脉内调制信号的瞬时频率,然后对信号瞬时频率进行直方图统计和局部峰值检测,提取了10个特征参数,实现了信号的识别。其中,采用了多重相位差分、信噪比阈值频率突变消除和载频粗估计等信号预处理方法,改善了算法对低信噪比环境的适应能力。计算机仿真表明,当信噪比高于4dB时,信号识别概率大于90%。硬件实现表明,信号识别所需时间小于0.2ms,算法具有一定的工程实用价值。

Recognition of intra-pulse modulation signal types based on phase difference

Based on phase difference, an approach for intra-pulse modulation recognition is presented. This approach can classify seven kinds of signals by ten characteristic parameters, which are picked up from the histogram and the peak value measurements of signal’s instantaneous frequency. Multi-phase differences, carrier frequency estimation and unwrapping frequency ambiguity caused by SNR threshold are used to improve the stability of this approach under the environment of low SNR. The simulation results show that the recognition rate can reach 90% when SNR is above 4dB. The implementation results in ADSP-TS201 whose clock operates at 600MHz show that less than 0.2 ms is cost when the signal’s pulse width is 2 μs. The simulation experiments and hardware implementations show the validity of this approach.