Weak signal detection based on stochastic resonance

In a heavy noise, the method based on Wigner Ville (WV) and Hough transformations has the poor performance of detecting the weak signal. To improve it, this paper analyzes the influence of the Fitz Hugh Nagumo (FHN) model's parameters on its filtering characteristics, and presents a method of detecting weak sinusoid and LFM signals based on FHN model of stochastic resonance. Firstly, the received signal is filtered by FHN model, and transformed by WV and Hough in turns, thus whether a signal is present in noise is determined according to whether there is a line in time-frequency picture. Finally, the validity of this method is well verified by the experiments.