Weak Signal Detection Based On Stochastic Resonance

The concept of Stochastic Resonance(SR) was first put forward during the process of revealing the changing regulation of climates in ancient time, and was defined as the phenomenon that adding appropriate noise can strengthen the system signal output SNR. Using the SR technology is obviously better than the tranditional signal detecting method. The best virtue of it is using the noise rather than restraining it.After giving the SR system equation, with the computer simulation and Runge-Kutta algorithm here, systematic research the relation is hit in common between noise signal and the nonlinearity system. Simulated result has indicated that the system improves the detecting ability of weak signal form the strong noise.Then in binary hypothesis testing problems, we will analyze the stochastic resonance (SR) effect of a fixed threshold using Newman-Pearson criteria. It is presented where performance comparisons are made between detectors where the Gaussian noise as well as, uniform, and symmetric bi-value noises are applied to enhance detection performance.Starting with the probability density function (PDF) of envelope Gaussian mixture (GM) model whereas the PDF of envelope by K-distribution model, we search the statistic rule between the parameter of them and SR system. After that, using the SR algorism we will detect the reverberation background signal and the result show that the detector is available.