Stochastic Resonance in the Optoelectronic Artificial Spiking Neuron Based on a VCSEL and a SPAD

The results of experimental observation of ghost stochastic resonance (GSR) in an optoelectronic artificial spiking neuron (ASN) based on a vertical-cavity laser and a singlephoton detector are presented. The ASN is a nonlinear threshold system that simulates the main properties of a biological neuron. It is excited by a subthreshold periodic signal consisting of a sum of ultraharmonics of a fundamental frequency, which is absent in the input signal itself. When noise of a certain amplitude is added to the input signal, a spectral component at the fundamental frequency appears in the spectra of the output time series. GSR is characterized by a nonmonotonic dependence of the response at the fundamental frequency on the amplitude of the added noise. The influence of the amplitude of the input subthreshold signal and the intensity of normally distributed noise on the manifestation of GSR is studied. The probability density function of the inter-spike interval distribution in the ASN's time dynamics, as well as the signal-to-noise ratio at the ghost frequency, were also investigated