A Novel Signal Denoising Method Based on Adaptive Stochastic Resonance

In non-cooperative communication, most of the received signals have low signal-to-noise ratio and less prior information. Therefore, before subsequent processing, the received signal should be effectively denoised. Denoising methods such as wavelet denoising and time-frequency transform have the problem of reducing noise and signal energy at the same time. Aiming at this problem, a large parameter variable scale adaptive tri-stable stochastic resonance (SR) method is proposed to denoise the multi-phase shift keying and multi-orthogonal amplitude modulation signals. On the basis of studying the traditional SR, the secondary sampling expansion frequency is adopted, and the fitness function based on singular value decomposition is used as the evaluation standard. The parameters of the tristable SR system are optimized by particle swarm optimization algorithm. The simulation results show that in the Gaussian channel environment, compared with the bi-stable adaptive SR method, this method can improve the output peak value and reduce the noise threshold to a certain extent, which is suitable for the enhancement of communication signals.