基于ELMAN网络的非线性混沌微弱信号检测

A Method for Nonlinear Chaos Weak Signal Detection Based on ELMAN Network

为了快速和实时地从具有强噪声的较低信噪比的原始信号中检测出有用信息,设计了一种混沌相空间重构理论和ELMAN神经网络的信号检测方法;首先,描述了采用混沌相空间重构理论对原始信号进行重构的原理和方法,在获取重构的时间序列的基础上,采用ELMAN网络来近似表示用于检测信号的函数型,然后,设计了ELMAN网络中各层之间连接权值的计算方式,并提出了采用ELMAN网络进行信号检测的具体过程,最后给出了采用混沌相空间重构理论和ELMAN网络的信号检测模型;对Lorenz混沌系统模型进行仿真实验,结果证明了文章方法能有效地对瞬时信号和周期性信息进行检测,在具有高斯白噪声的情况下,仍然具有降噪效果好的优点,是一种用于信号检测的可行性方法。

In order to detect the useful information in the primitive signal with strong noise quickly and in time,a signal detection method based on chaos phrase reconstruction theory and ELMAN network is proposed,Firstly,the principle and method based on chaos phase space reconstruction theory is described.Then the ELMAN network is used to approximately represent the function type on the obtained reconstruction time sequence,then the connecting weight for computing the weight among layers is designed,and the ELMAN network for detecting the signal is proposed.The simulation experiment result shows：the method in this paper can detect the momentary signal and cyclic signal,under the circumstance of Gaussian white noise,still having the good decomposition effect,so it is a feasible method for signal detection.