基于并联自适应随机共振的微弱信号检测方法

发布时间：2018-11-05 16:43

【摘要】：针对传统大参数信号自适应随机共振方法存在的变换尺度变化范围选取缺乏固定标准、参数自适应效率低、检测到的目标信号不够明显等不足,提出一种基于并联自适应随机共振的微弱信号检测方法,实现强噪声背景下大参数微弱信号的快速、有效检测。推导出基于采样频率的变换尺度的最大变化范围,将该范围平均分段,以输出信噪比为适应度函数,在变换尺度各子搜索范围和共振系统参数搜索范围内,采用带极值扰动的简化粒子群算法对变换尺度及系统参数进行自适应优化选择;将优化得到的变换尺度和系统参数分别作为并联各子随机共振系统的输入信号变换尺度和系统参数;将各子系统的输出响应分别进行自相关处理后,合成为最终的系统输出响应。仿真及实际应用结果表明,该方法对强噪声背景中的微弱目标信号具有准确的检测能力,能够有效提高参数自适应效率,突出检测到的目标信号,增强强噪声背景下微弱信号的检测能力。
[Abstract]:In view of the shortcomings of traditional adaptive stochastic resonance (SRR) methods, such as the lack of a fixed standard for the selection of the range of transformation scales, the low efficiency of parameter adaptation, and the lack of obvious detected target signals, etc. A method of weak signal detection based on parallel adaptive stochastic resonance is proposed, which can detect weak signal with large parameter under strong noise quickly and effectively. The maximum variation range of the transform scale based on sampling frequency is deduced. The range is divided into sections, and the output signal-to-noise ratio (SNR) is taken as the fitness function. In the sub-search range of the transform scale and the parameter search range of the resonance system, A simplified particle swarm optimization algorithm with extremum perturbation is used to optimize the transformation scale and system parameters. The optimized transformation scale and system parameters are used as input signal transformation scale and system parameters of parallel sub-stochastic resonance system respectively. The output response of each subsystem is synthesized into the final output response after autocorrelation. The results of simulation and practical application show that the method has accurate detection ability for weak target signal in strong noise background and can effectively improve the parameter adaptive efficiency and highlight the detected target signal. Enhance the detection ability of weak signal in strong noise background.
【作者单位】：海军航空工程学院7系;新疆军区指挥自动化站;
【基金】：国防预研基金项目(9140A27020214JB14435)
【分类号】：TN911.23

【相似文献】