Random noise in the spectra of evoked otoacoustic emissions

Noise from the measuring equipment and from environmental and physiological sources is inevitably present in records of evoked otoacoustic emissions. This paper considers the influence of noise on the power spectra of otoacoustic emissions as recorded with the Otodynamics ILO88 system. It is shown that if the noise has a normal distribution, the ratio of the power in the response spectrum to the power in the noise has an approximately log-normal distribution. It is also shown that the change in the level of the response spectrum that is observed when measurements are repeated is related through a t-distribution to the estimate of the noise-to-signal ratio that these measurements provide. The statistical analysis leads to simple significance tests that may be helpful in detecting the presence of an emission and that allow small changes in the spectrum to be identified.