带限噪声下基于Duffing振子的信号检测方法研究

发布时间：2018-03-21 01:51

本文选题：信号检测　切入点：Duffing振子　出处：《哈尔滨工业大学》2017年硕士论文　论文类型：学位论文

【摘要】：信号检测是通信领域的重要环节,无论是合作方抑制频偏的信号接收还是非合作方扫描信号的信号检测,都需要先准确地检测到信号的载波频率。而现代通信要求在更低的信噪比条件下,利用更少的先验知识,检测更复杂的信号,为此人们提出利用非线性方法代替传统的线性方法实现低信噪比的信号检测。非线性方法对噪声的抵抗能力强,能够在更低的信噪比条件下检测信号,但是目前还停留在理论阶段,没有实际应用。Duffing振子作为混沌理论的典型代表是一个很好的非线性系统,能够用来检测信号的载波频率,但是传统的基于Duffing振子的信号检测方法都是在宽带噪声的基础上研究的,而且只关注了一个信号的问题。为此,本论文以高功率谱密度带限高斯噪声为基础,利用单音信号模拟信号之间的干扰,研究基于Duffing振子的新的信号检测方法,以适合更加普遍的通信环境。本论文首先研究了Duffing振子的基本原理,对差分振子进行分析,得到更合适的参数及阈值选择方法。然后分析了高功率谱密度带限噪声对传统的Duffing振子检测信号方法的影响,验证了Duffing振子对噪声和单音干扰的抵抗能力。进而将噪声带入到Duffing振子的模型中,得到了在高功率谱密度带限噪声条件下的Duffing振子新状态,理论推导了Duffing振子周期状态出现概率随策动力幅值初始值的变化规律,并重新定义了广义周期状态和广义混沌状态,通过检测PSTR(周期状态持续时间比重)稳定值判断Duffing振子状态。随后在此基础上提出了广义间歇混沌现象及理论,并建立了基于广义间歇混沌的信号检测方法。最后利用MATLAB设计了信号检测平台,验证了在高功率谱密度带限噪声条件下,基于广义间歇混沌的信号检测方法在单信号和多信号检测中的可行性,并通过与其他检测方法的比较,验证了该方法的优越性。
[Abstract]:Signal detection is an important link in the field of communication. Modern communications require less prior knowledge to detect more complex signals under lower signal-to-noise ratio (SNR) conditions. Therefore, the nonlinear method is proposed to replace the traditional linear method to detect the signal with low SNR. The nonlinear method has strong resistance to noise and can detect the signal under the condition of lower SNR. But at present, it is still in the theoretical stage, and no practical application of .duffing oscillator as a typical representative of chaotic theory is a good nonlinear system, which can be used to detect the carrier frequency of the signal. However, the traditional signal detection methods based on Duffing oscillator are studied on the basis of wideband noise, and only one signal problem is concerned. Therefore, this paper is based on high power spectral density band-limited Gao Si noise. A new signal detection method based on Duffing oscillator is studied by using the interference between the analog signals of single tone signal, which is suitable for more general communication environment. In this paper, the basic principle of Duffing oscillator is studied, and the differential oscillator is analyzed. A more suitable parameter and threshold selection method is obtained, and then the influence of high power spectral density band-limited noise on the traditional Duffing oscillator detection signal method is analyzed. The ability of Duffing oscillator to resist noise and monophonic interference is verified. Furthermore, the noise is introduced into the model of Duffing oscillator, and a new state of Duffing oscillator is obtained under the condition of high power spectral density band-limited noise. The variation of the occurrence probability of the periodic state of the Duffing oscillator with the initial value of the amplitude of the excitation force is derived theoretically, and the generalized periodic state and the generalized chaotic state are redefined. The state of Duffing oscillator is judged by detecting the stable value of PSTR (periodic state duration specific gravity). Based on this, the generalized intermittent chaos phenomenon and its theory are proposed. A signal detection method based on generalized intermittent chaos is established. Finally, a signal detection platform is designed by using MATLAB, which verifies that under the condition of high power spectral density band-limited noise, The feasibility of signal detection method based on generalized intermittent chaos in single signal detection and multi-signal detection is verified by comparing with other detection methods.
【学位授予单位】：哈尔滨工业大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN911.23

【参考文献】