The Analysis of the Quasi-likelihood Algorithm of the Estimated Parameter of the Signal with Unknown Localization within the Observation Interval

摘要

We carried out the synthesis and analysis of the quasi-likelihood estimation algorithm of regular parameter of an untailored pulse with the unknown moments of appearance and disappearance observed against additive Gaussian white noise. We studied the influence of prior ignorance of the moments of appearance and disappearance on accuracy of the introduced estimate. As an example we considered the quasi-likelihood estimates of the amplitude and initial phase of a rectangular radio signal. And we found the loss in estimation accuracy due to ignorance of the moments of appearance and disappearance.