阵列级联随机共振原理及其在分布式融合检测中的应用

The demand for the acquisition of low observable object information on the condition of extremely low SNR is much higher than ever before, and the detection threshold can hardly be reduced further by the traditional detection method. Stochastic resonance phenomenon indicates that under certain conditions noise energy can transform to signal energy, enhance the output SNR as well as the order of the system, and thus improve the detection probability of weak signals. The key issue of stochastic resonance and its application in the detection of weak signals is to understand the inner mechanism of stochastic resonance induced by the coupling effects of the system, the signal and the noise..Considering the existing research techniques are mostly restricted to numerical simulation, this project expects to introduce methods such as stochastic linearization to establish approximate mathematic models in order to transform the nonlinear systems, which is difficult to obtain an accurate solution, to equivalent linear or nonlinear systems, which can be mathematically analyzed more easily, and furthermore instigate the effects of the systematic parameters and additive noise on the stochastic resonance phenomenon. On this bases, this project will analyze the relationships between the array cascaded stochastic resonance and the distributed fusion detection method, deduce the optimal detection criterion, and accordingly establish detection method on the condition of extremely low SNR. The implementation of this project can provide significant theoretic guidance to the thorough understanding of the inner mechanism of stochastic resonance as well as its enhanced application in detection fields for weak signals.

低可观测目标信息的获取对极低信噪比条件下微弱信号检测性能的要求越来越高，传统检测方法已很难进一步降低检测门限。随机共振现象表明在一定条件下噪声能量可转换为信号能量，从而增强系统输出的信噪比和有序性，提高微弱信号检测概率。但随机共振及其在微弱信号检测中应用的关键问题在于揭示系统、信号、噪声耦合作用产生随机共振现象的内在机制。.针对已有研究手段大多局限于数值仿真的现状，本项目拟采用随机线性化等方法将难于求得精确解的非线性系统转化为在数学上更易于分析的等效系统，建立近似定量模型，揭示系统参数、外加噪声对阵列级联随机共振的作用影响和作用方式。在此基础上，本项目将研究阵列级联随机共振与分布式并联反馈融合检测方法的作用关系，推导最优检测准则，据此建立极低信噪比条件下的微弱信号检测方法。本项目的顺利开展可为更全面、深入地认知随机共振现象的内在机制，加强随机共振在微弱信号检测领域的应用提供重要的理论指导。

低可观测目标信息的获取对极低信噪比条件下微弱信号检测性能的要求越来越高，传统检测方法已很难进一步降低检测门限。随机共振现象表明在一定条件下噪声能量可转换为信号能量，从而增强系统输出的信噪比和有序性，提高微弱信号检测概率。本项目针对低可观测目标信息获取问题，研究了系统参数、外加噪声等对阵列级联随机共振的影响机制，以及基于随机共振原理和分布式融合检测的微弱信号方法。.在随机共振理论研究方面，针对典型分数阶阵列级联动力学系统，研究了系统参数、噪声参数、耦合强度等对系统随机共振现象的影响规律，对深入认知复杂系统随机共振机理具有重要意义。在微弱信号检测方法研究方面，提出了新的耦合乘积型指数滤波器和样本不足情况下的机载MIMO雷达协方差估计方法，有效提高目标检测能力，为发展新的微弱信号检测方法奠定了基础。相关研究成果已在国防科技研究项目中得到了应用，具有广阔的应用前景。