Signal Detection Study Of Adaptive Stochastic Resonance Based On PSO Algorithm

When the stochastic resonance phenomenon satisfy the conditions of stochasticresonance show their unique advantages in signal processing. For example, in thefield of the weak signal detection and processing, compared with the traditionalvariety of signal detection methods, the stochastic resonance is not to reduce the noise,but useless "noise" is used to strengthen a weak signal. Therefore, using stochasticresonance theory combined with optimization algorithm to implement adaptivestochastic resonance in signal processing, testing and other science research haveplayed an important role. Based on the signal detection in concurrent evolution ofmultiple parameter stochastic resonance and got a better to improve the output signal.PSO algorithm is a typical intelligent optimization algorithms that simulate thebehavior of bird populations in search of food and conceived algorithms. It has asimple algorithm, easy to implement, adaptability and other characteristics. Moreimportant is its multi-objective search capabilities, global synchronization instochastic resonance can be quickly and effectively realize adaptive characteristics ofstochastic resonance.Firstly，this paper has summarized the theory of stochastic resonance andrelevant theoretical knowledge as well as several types, and through the simulation,the output of the observation system signal-to-noise ratio and the change of noiseintensity D relations.Secondly, in view of the traditional stochastic resonance methodto system parameters and noise intensity of matching for the limitations of theresearch background, in order to solve the problems of the bistable system parametersreasonable selection, and to overcome the traditional adaptive stochastic resonance ofthe single parameter optimization, the deficiency of the two kinds of parametersselection algorithm is given. Apply the basic particle swarm algorithm to implementadaptive stochastic resonance (at the same time, optimize the system structure parameters), and the output signal simulation experiment was carried out.Finally, the improved PSO algorithm is proposed, in which multiple adaptivestochastic resonance method of signal detection. The method system of the outputsignal-to-noise ratio as the optimization objective function, the improved particleswarm algorithm has strong global search ability and can maintain the diversity ofparticles () to optimize parameters. The improved PSO algorithm is applied to themulti-layer bistable stochastic resonance in the optimization of parameters, the bestworking state of the stochastic resonance system. The simulation results show that themethod can improve the output SNR, the algorithm converges fast, testing effect isgood.