White noise analysis of nonlinear systems with application to the auditory system

Authors

  • J.J. Eggermont Dept. of Psychol., Calgary Univ., Alta., Canada
  • G.M. Smith Dept. of Psychol., Calgary Univ., Alta., Canada

Keywords:

hearing, nonlinear systems, physiological models, auditory system, cross correlation, Gaussian wide band noise, uniform distributed random number generator, grassfrog

Abstract

Linear systems analysis with tones, clicks or white noise results essentially in the same information: the impulse response or frequency response of the system. In non-linear systems, such as the auditory system, these three methods provide different results. Furthermore the outcome will be level dependent. Higher order cross correlation with Gaussian wide band noise as input signal provides, in principle, a method to analyze non-linear systems. For that purpose one needs noise with a zero valued second order auto correlation function. Commercially available pseudo-random noise generators do not produce noise with satisfactory properties. Software generated noise with Gaussian amplitude distribution can easily be generated on basis of the uniform distributed random number generator. Using noise with these improved characteristics the authors studied neurons in the auditory midbrain of the grassfrog. Two examples are shown and a comparison is given between the results obtained with the white noise method and the more traditional pure tone analysis

Additional Files

Published

1988-03-01

How to Cite

1.
Eggermont J, Smith G. White noise analysis of nonlinear systems with application to the auditory system. Canadian Acoustics [Internet]. 1988 Mar. 1 [cited 2025 Feb. 20];16(1):3-15. Available from: https://jcaa.caa-aca.ca/index.php/jcaa/article/view/584

Issue

Section

Technical Articles