In this paper,the effects of a bistable potential function U(x)= ax 2 /2 + b|x| 2γ /(2γ) on stochastic resonance (SR) is discussed.We investigate the effects of index γ on the performance of the SR system with fixed parameters a and b,and with fixed potential barriers,respectively.To measure the performance of the SR system in the presence of an aperiodic input,the bit error rate is employed,as is commonly used in binary communications.The numerical simulations strongly support the theoretical results.The goal of this investigation is to explore the effects of the shape of potential functions on SR and give a guidance of nonlinear systems in the application of information processing.
In this paper,we discuss the effects of error feedback on the output of a nonlinear bistable system with stochastic resonance.The bit error rate is employed to quantify the performance of the system.The theoretical analysis and the numerical simulation are presented.By investigating the performances of the nonlinear systems with different strengths of error feedback,we argue that the presented system may provide guidance for practical nonlinear signal processing.
This paper presents a novel approach of M-ary baseband pulse amplitude modulated signal processing via a parameter-optimized nonlinear dynamic system.This nonlinear system usually shows the phenomenon of stochastic resonance by adding noise.To thoroughly discuss the signal processing performance of the nonlinear system,we tune the system parameters to obtain a nonlinear detector with optimal performance.For characterizing the output of the nonlinear system,the derivation of the probability of detection error is given by the system response speed and the probability density function of the nonlinear system output.By varying the noise intensity with fixed system parameters,the phenomenon of stochastic resonance is shown and by tuning the system parameters with fixed noise,the probability of detection error is minimized and the nonlinear system is optimized.The detection performance of the two cases is compared with the theoretical probability of detection error,which is validated by numerical simulation.
针对不确实海洋环境使声呐探测性能下降和宽容性差的问题,提出了环境参量和信号参量不确实性两嵌入的宽容波束形成贝叶斯方法.不确实环境参量的先验概率密度函数(probability density function,PDF)通过多项式混沌展开与传播模型相结合嵌入到接收信号的概率建模中,导出信号参量的先验PDF;接收信号参量先验PDF通过贝叶斯波束形成嵌入到处理中,转化为后验PDF.导出的后验概率最大贝叶斯波束形成具有估计器和相关器相结合的GLRT结构.仿真和实验数据分析结果表明:不确实环境参量导致接收声场发生秩扩展,由秩1扩展为秩2或秩3,贝叶斯波束形成体现了相干匹配与非相干积累的结合,实现了浅海环境中目标的正确定位,增加了宽容性.
