On the active periods of nonlinear shot noise

Iddo Eliazar*, Joseph Klafter

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

5 Scopus citations

Abstract

We study the temporal 'on-and-off' structure of the recently introduced nonlinear Shot Noise model [I. Eliazar, J. Klafter, PNAS 102 (2005) 13779; I. Eliazar, J. Klafter, Nonlinear shot noise: Lévy, Noah, & Joseph, Physica A (2006), to appear], in which: (i) shots of random magnitudes 'bombard' a system stochastically in time; (ii) the magnitudes of 'incoming' shots decay to zero nonlinearly; and, (iii) the overall effect of the shots on the system is additive. When the shot-inflow and shot-decay satisfy certain conditions, the resulting Shot Noise alternates randomly between 'active periods' (in which the noise is 'on') and 'silent periods' (in which noise is 'off'). The statistical properties of the 'active' and 'silent' periods are analyzed, and the resulting 'on-and-off' Shot Noise structure is explored. Explicit formulae for means, variances, Laplace transforms, and other statistics are derived in closed-form. Based on this analysis, Shot Noise systems are categorized into three classes: transient, null-recurrent, and recurrent; and, these classes are characterized both analytically and probabilistically. Within the null-recurrent class, systems whose 'active periods' are governed by heavy-tailed probability distributions are further characterized.

Original languageEnglish
Pages (from-to)237-259
Number of pages23
JournalPhysica A: Statistical Mechanics and its Applications
Volume363
Issue number2
DOIs
StatePublished - 1 May 2006

Keywords

  • First Passage Times
  • Heavy-tailed distributions
  • Nonlinear Shot Noise
  • Nonlinear decay
  • Poisson inflow

Fingerprint

Dive into the research topics of 'On the active periods of nonlinear shot noise'. Together they form a unique fingerprint.

Cite this