基于高阶统计量的多模噪声中的信号检测

High-Order Statistics-based Signal Detection In Multi-modal Noise

按照概率密度函数形状,给出了一种比较通用的非高斯噪声模型——多模噪声。多模噪声总体上属于非高斯噪声,但兼容了高斯噪声。改进了高阶统计量的双谱算法,给出一种基于双谱的多模噪声中信号的检测方法,并在此基础上结合无惯性非线性变换器和双谱技术,改进了传统的自适应幅频干扰抑制器,可以精确估计或检测信号。仿真表明该方法可以抑制高斯噪声,同时在强噪声和复杂背景下可以以较高的检测概率检测出信号,优于传统的似然比检测。

In accordance with the shape of probability density function,a more universal model of non-Gaussian noise known as multi-modal noise is defined.Multi-modal noise,as a whole,is non-Gaussian,but comprises Gaussian noise.The bispectrum algorithm of high-order statistics comprising Gaussian noise is improved,and the method for signals detection in multi-modal noise given.On this basis,the self-adapting amplitude and frequency interference suppresser,in combination of inertialess nonlinear transformer and bispectrum is designed.It could exactly estimate and detect signal.This method could suppress Gaussian noise and with higher detection probability,detect the signal in strong-noise and complex background.Simulation results show that it is superior to the traditional likelihood-ratio dection.