基于Duffing振子的自适应随机共振微弱信号检测

发布时间：2018-06-29 19:37

本文选题：随机共振 + Duffing混沌振子　；参考：《南京信息工程大学》2017年硕士论文

【摘要】：在精密仪器测量中,信号往往会被各类噪声所覆盖,利用随机共振特性,精确提取强噪声背景下的微弱特征信号信息,实现极低信噪比状态下的微弱信号检测,成为近年来研究的热点之一。本文以二维Duffing振子为随机共振发生载体,研究了 Duffing混沌系统各参数诱导随机共振问题,分别提出了基于全局人工鱼群算法的自适应随机共振微弱信号检测和基于Duffing振子随机共振和遗传算法的频率调制微弱信号检测方法,建立了检测模型,通过仿真实验,验证了检测效果。具体研究如下:从随机共振线性响应和绝热近似理论角度,推导了随机共振产生及其限制条件。基于Duffing振子随机共振特性,采用手动设置参数方法,实现了 α稳定噪声背景下随机共振微弱信号检测;在高斯白噪声背景下,研究了双稳态系统参数对随机共振模型结构的影响;在α稳定噪声背景下,研究了二维Duffing振子系统各参数对随机共振输出的影响,为自适应随机共振系统搭建提供了理论支持。为克服传统自适应随机共振单一参数优化的缺陷,解决多参数调节问题,提出基于全局人工鱼群算法的自适应随机共振微弱信号检测。以系统全参数为优化对象,比较了二维Duffing振子随机共振和一维Langevin随机共振的自适应微弱信号检测系统。结果表明:在相同条件下,二维Duffing随机共振自适应系统更具优越性,并将基于Duffing振子的自适应随机共振系统成功应用于海杂波背景下的微弱信号检测,提升了海杂波背景下的小目标信号检测性能。针对随机共振微弱信号检测范围有限、受小参数限制问题,根据频率调制原理,提出基于二维Duffing随机共振和遗传算法的频率调制微弱信号检测。数值分析和仿真结果表明:所提方法灵活性强,模型鲁棒性好,能够有效从强噪声背景下提取微弱特征信号,不仅适用于高、低频微弱信号检测,还适用于多频微弱信号的检测。此研究扩展了随机共振微弱信号检测范围,为实际工程中的Duffing振子随机共振微弱信号检测提供依据。
[Abstract]:In the precision instrument measurement, the signal is often covered by various kinds of noise. Using the stochastic resonance characteristic, the weak characteristic signal information under the strong noise background is extracted accurately, and the weak signal detection under the extremely low signal-to-noise ratio is realized. It has become one of the hot research topics in recent years. In this paper, the problem of induced stochastic resonance of duffing chaotic system is studied by using two-dimensional duffing oscillator as the carrier of stochastic resonance. Adaptive stochastic resonance weak signal detection based on global artificial fish swarm algorithm and frequency modulation weak signal detection method based on duffing oscillator stochastic resonance and genetic algorithm are proposed respectively. The detection effect is verified. The main results are as follows: from the point of view of stochastic resonance linear response and adiabatic approximation, the generation of stochastic resonance and its limiting conditions are derived. Based on the stochastic resonance characteristics of duffing oscillator, the method of manually setting parameters is used to detect the weak signal of stochastic resonance in the background of 伪 stable noise, and in the background of Gao Si white noise, The influence of the parameters of bistable system on the structure of stochastic resonance model is studied, and the influence of the parameters of two-dimensional duffing oscillator system on the output of stochastic resonance is studied under the background of 伪 -stable noise, which provides theoretical support for the construction of adaptive stochastic resonance system. In order to overcome the shortcomings of traditional adaptive stochastic resonance single parameter optimization and solve the problem of multi-parameter adjustment, an adaptive stochastic resonance weak signal detection method based on global artificial fish swarm algorithm is proposed. The adaptive weak signal detection system with two-dimensional duffing oscillator stochastic resonance and one-dimensional Langevin stochastic resonance is compared. The results show that under the same conditions, the two-dimensional duffing stochastic resonance adaptive system is superior, and the adaptive stochastic resonance system based on duffing oscillator is successfully used to detect the weak signal in the background of Yu Hai clutter. The performance of small target signal detection in sea clutter background is improved. Aiming at the problem that the detection range of stochastic resonance weak signal is limited and limited by small parameters, according to the principle of frequency modulation, a frequency modulated weak signal detection method based on two-dimensional duffing stochastic resonance and genetic algorithm is proposed. Numerical analysis and simulation results show that the proposed method is flexible and robust, and can extract weak feature signals from strong noise background effectively, which is not only suitable for detection of high and low frequency weak signals. It is also suitable for the detection of multi-frequency weak signals. This study extends the detection range of stochastic resonance weak signal and provides the basis for duffing oscillator random resonance weak signal detection in practical engineering.
【学位授予单位】：南京信息工程大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN911.23

【参考文献】