随机共振在微弱信号检测中的数值仿真研究

【作者】 张军；

【导师】 王辅忠；

【作者基本信息】 天津工业大学 ， 物理电子学， 2007， 硕士

【摘要】 随机共振是噪声和周期信号作用于非线性系统，信号和噪声在非线性系统的协同作用下，发生噪声能量向信号转移产生的一种现象。利用随机共振原理可提高系统输出高信噪比达到识别弱信号的目的。该方法与常规线性滤噪方法相比，随机共振不是消除噪声而是充分利用噪声能量来放大信号提高输出信噪比。目前，已有的随机共振绝热近似理论仅适用于周期性弱信号(幅度、频率和噪声强度均小于1)，为了在实际大参数下仍然能用绝热近似理论得到随机共振现象，本课题对大频率的周期信号检测进行了初步的研究和探索。本文探讨了双稳态系统势垒与系统参数关系，系统地研究了工程测量中常见的噪声、信号和非线性系统之间的关系以及产生随机共振的最佳条件，对微弱信号的检测方法进行了深入的研究。应用计算机仿真技术，基于Runge-Kutta算法，提出了频率变换的随机共振方法，以实现强噪声背景下大频率周期信号的检测。仿真结果表明，此方法提高了强噪声背景下微弱信号的检测能力。随机共振技术在信号检测方面有着潜在的应用价值，有望将来应用于实测信号的数据处理。如何将随机共振技术应用到实际信号的检测还需要做大量的研究工作。更多还原

【Abstract】 Stochastic resonance is that noise and the periodic signal act on non-linear system, one kind of phenomenon that the noise energy produces to signal metastasis happened in the signal and noise under non-linear system coordination effect. The signal-to-noise ratio making use of stochastic resonance principle but improving system output height reaches the purpose distinguishing the weak signal. The method is compared with the routine linear filters; the stochastic resonance is not to remove noise but is that noise enlarges the signal energy improving the signal-to-noise ratio of output fully. At present, the similar theory of adiabatic elimination applies to the weak signal of periodic only (amplitude, frequency and noise intensity are smaller than 1 equally). For still being able to use similar theory of adiabatic elimination to get stochastic resonance phenomenon under big actual parameters, the problem has carried out the preliminary research and exploration on high frequency periodic signal detection.Computer emulation technology the main body of a book is applied, systematic research engineering survey is hit by common the relation between noise, signal and nonlinearity system and the best condition producing stochastic resonance , method carries out the preliminary research and exploration on high frequency signal detection. With the computer simulation and Runge-Kutta algorithm here, the new method of the adiabatic elimination SR principle for detecting a weak signal of high frequency is analyzed from the strong noise. Simulated result has indicated that the method improves the detecting ability of weak signal from the strong noise.The stochastic resonance technology has latent application value, hopeful with the future in the field of signal detection apply to practical measures of the signal processing. The detecting how to apply stochastic resonance technology to actual signal needs to do large amount of job.更多还原