Statistical characterization of time variability in midlatitude single-tone HF channel response

Abstract

In this paper, a statistical analysis approach is proposed to characterize the variability of HF channel response to single-tone signals by using only the amplitude information of the received signal. By the proposed methodology, robust estimates of the time-varying mean and variance of the channel response can be obtained. For this purpose, we use sliding window statistics of the available data. On the basis of the estimated variance of the obtained results, a detailed justification of the proper window size is given. In order to obtain more reliable estimates, the data are median filtered prior to statistical analysis. A robust way of choosing the length of the median filter is presented. We applied the statistical analysis approach to a set of available data obtained from a measurement campaign between England and Turkey conducted from April 1992 to February 1993. The results of the statistical analysis confirmed the expectations of the physical behavior of the ionospheric channel. It was found that the midlatitude single-frequency channel is slowly time-varying and locally stationary in a sliding window of 22 s. Also, it was observed that the amplitude of the received signal exhibits a significant diurnal variation. In addition, during early morning hours and night hours, the channel is considerably more stable for communication purposes compared with day and early evening hours.