【摘要】 在电子对抗领域,信号调制方式识别是进行雷达分选、干扰施放的基础,得到广泛研究。对此,文中提出了一种以信号频谱相像系数和幅度统计参数为分类特征的FSK/BPSK复合调制雷达脉冲信号识别算法。算法首先提取雷达脉冲信号的频谱相像系数和幅度统计参数,然后采用分层结构的神经网络分类器进行识别。该算法不仅能识别FSK/BPSK复合调制信号,且对其他常用雷达信号调制方式的识别不产生干扰。仿真结果表明,针对FSK/BPSK以及CW、LFM、BPSK、QPSK、FSK等常用雷达信号调制类型,在信噪比>5 d B时,分类正确率可达98%以上。更多还原

【Abstract】 The recognition of signal modulation is a base problem for sorting radar signals and releasing jamming signals in the area of electronic countermeasure. In this paper,a novel method is proposed for the recognition of hybrid modulation radar signal combined with FSK and BPSK based on the resemblance coefficient and amplitude parameter. The proposed approach recognizes FSK / BPSK signals without negative effect on the modulation recognition of other received radar signals. After feature extracting of radar signal spectrum,a hierarchical neural network is designed for identifying the common used six radar modulation types. Simulation results show that the successful recognition rate is over 98% with the Signal-to-Noise-Ratio( SNR) greater than 5 d B.更多还原