Nonstationary Stochastic Resonance

TL;DR

This paper demonstrates that stochastic resonance, the phenomenon where noise enhances weak signal detection, extends to nonstationary signals, broadening its potential applications in biological and electronic systems.

Contribution

It introduces a multilevel trigger mechanism for nonstationary stochastic resonance, showing optimal detection with around ten thresholds, and explores its biological relevance.

Findings

Optimal detection occurs at about ten thresholds.

Stochastic resonance applies to nonstationary signals.

Potential evolutionary link to fixed sensory thresholds.

Abstract

It is by now established that, remarkably, the addition of noise to a nonlinear system may sometimes facilitate, rather than hamper the detection of weak signals. This phenomenon, usually referred to as stochastic resonance, was originally associated with strictly periodic signals, but it was eventually shown to occur for stationary aperiodic signals as well. However, in several situations of practical interest, the signal can be markedly nonstationary. We demonstrate that the phenomenon of stochastic resonance extends to nonstationary signals as well, and thus could be relevant to a wider class of biological and electronic applications. Building on both nondynamic and aperiodic stochastic resonance, our scheme is based on a multilevel trigger mechanism, which could be realized as a parallel network of differentiated threshold sensors. We find that optimal detection is reached for a number of thresholds of order ten, and that little is gained by going much beyond that number. We raise the question of whether this is related to the fact that evolution has favored some fixed numbers of precisely this order of magnitude in certain aspects of sensory perception.