基于Duffing振子的MSK信号检测与识别技术研究

发布时间：2018-03-14 03:46

本文选题：Duffing振子　切入点：MSK调制　出处：《长春理工大学》2017年硕士论文　论文类型：学位论文

【摘要】：在信息传输与信息对抗中,通信信号的检测与调制识别是十分重要的环节。为了提高通信信号的功率效率和安全性,使用诸如MSK、GMSK、扩谱等现代调制技术的通信信号,由于这类通信信号功率谱密度较低,在强噪声的环境下,传统的信息对抗方法很难进行检测和识别。作为非线性科学分支的混沌理论,由于其具有对噪声免疫以及对微弱信号敏感的特点,在微弱信号检测领域具有极大的优越性。本文提出利用Duffing振子进行MSK信号的检测与调制方式识别的研究。利用MSK调制信号相位连续、调制频率正交、频差最小的特点,首先建立了MSK信号的双Duffing振子检测模型,然后利用功率谱熵进行状态量化判定,并设计了信号的检测流程。最后,提出了基于四个Duffing振子组合阵列的调制信号识别系统,根据不同的调制信号使系统的状态发生不同组合变化原理,以功率谱熵作为特征,设计了基于RBF神经网络的分类器进行分类识别。该方案可识别2ASK、BFSK、BPSK和MSK已调信号。仿真结果表明,本文提出的检测与识别方法具有可行性。与传统的检测和分类识别方法相比,本文所提的技术方案具有检测灵敏、识别正确率高、对噪声免疫,且不依赖于先验信息的优势,在信噪比低至-20dB的条件下,系统仍然可以保持稳定工作,具有稳健性。
[Abstract]:In order to improve the power efficiency and security of communication signals, modern modulation technologies such as MSK / GMSK, spread spectrum and so on are used to detect and identify communication signals. Because of the low power spectral density of this kind of communication signal, it is difficult to detect and recognize the traditional information countermeasure method in the environment of strong noise. Because it is immune to noise and sensitive to weak signals, It has great superiority in the field of weak signal detection. In this paper, Duffing oscillator is used to detect and identify MSK signal. The characteristics of MSK modulation signal are continuous phase, quadrature modulation frequency and minimum frequency difference. First, the dual Duffing oscillator detection model of MSK signal is established, then the state quantization is determined by power spectrum entropy, and the signal detection flow is designed. Finally, a modulation signal recognition system based on four Duffing oscillator combination arrays is proposed. According to different modulation signals, the state of the system is changed by different combinations, which is characterized by power spectrum entropy. A classifier based on RBF neural network is designed to recognize 2ASK BFSKK and MSK modulated signals. The simulation results show that the proposed method is feasible and compared with the traditional detection and classification methods. The proposed scheme has the advantages of sensitive detection, high recognition accuracy, immunity to noise, and independent of prior information. Under the condition of low SNR to -20 dB, the system can still work stably and has robustness.
【学位授予单位】：长春理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN911

【相似文献】