Research On Weak Signal Detection Based On Stochastic Resonance

In the field of signal processing, when the nonlinear system is stimulated by signal and noise, the system’s best output performance can be achieved by adjusting the noise intensity and system parameters. In other words, the phenomenon of nonlinear system, the input signal and the input noise results in a synergistic effect which is called stochastic resonance(SR).The main goal of this thesis is using SR to detect weak period signal from strong-noise-background. Also the parameter-induced SR, noise-induced SR and mathematical model of SR are studied in detail. The thesis’ s arrangements and main contents are as following:Chapter one expatiate the background and significance of the research deeply, introduce a variety of conventional method of WSD. Then, the research status of WSD at home and abroad based on SR is introduced. Finally, the main contents of this thesis are also given.In chapter two, the mathematical model based on SR is studied deeply. Several kinds of important SR theories and several measure indexes of SR are introduced in detail, and the measure index of subsequent simulation experiments in this thesis is confirmed.In chapter three, the big frequency SR based on twice sampling is given, which is used to detect the weak signal not meeting the adiabatic approximation theory. The experiments in bistable system and monostable system are simulated respectively, and the simulation results are analyzed in detail.In chapter four, adaptive unknown big frequency SR based on frequency control is given. This method not only can achieve the large parameter SR, but also detect the unknown frequency signal. Its detection principle is that the unknown signal is modulated with control frequency firstly, and then the modulated signals which contain a new high frequency and a new low frequency are inputted into the SR system, the maximum signal-to-noise ratio is chosen as the measure index and the corresponding control frequency can be found, the frequency of unknown signal can be calculated at last.Chapter five builds a power function type monostable SR under Levy noise. The monostable system model is extended to more general power function type monostable system model, and the Gaussian noise is replaced by more complex Levy noise. Besides, the form of potential function of power function type monostable system and principle of Levy noise are described in detail. The laws for the resonant output of monostable system, governed by system parameters and the intensity amplification factor of Levy noise, are explored under different values. These results contribute to reasonably choosing the system parameters and intensity amplification factor of power function type monostable SR system under Levy noise, and provide a reliable basis for practical engineering application of WSD by SR.Chapter six reviews the full thesis, summarizes the main contents of this thesis, and discusses further research still to go.