Auto-covariance estimation of variable samples (ACEVS) and its applicationfor monitoring random ionospheric disturbances using GPS

Ionospheric variation may be considered as a stationary time series under q uiet conditions. However. the disturbance of a stationary random process fr om stationarity results in the bias of corresponding samples from the stati onary observations, and in the change of statistical model parameters of th e process. From a general mathematical aspect, a new method is presented fo r monitoring ionospheric variations, based on the characteristic of time-se ries observation of GPS, and an investigation of the statistical properties of the estimated auto-covariance of the random ionospheric delay when chan ging the number of samples in the time series is carried out. A preliminary scheme for monitoring ionospheric delays is proposed.