Free space optical (FSO) channel estimation and tracking using the incremental adaptive networks

Channel estimation is necessary for almost every communication system. Its importance becomes clear when the links between the transmitter and receiver become highly unstable. The channels of the free space optical (FSO) communication systems suffer from optical turbulence and there are several statistical models to explain this phenomenon. In this paper, we consider the FSO channel estimation with adaptive incremental networks and examine their performance in estimating the channels modeled with the Log-normal and Gamma-Gamma distributions. This is the first attempt to estimate FSO channels with adaptive shames and the results show that as the turbulence level of the considered FSO channel rises, its estimation becomes harder and with more error. Therefore, estimating Log-normal modeled FSO channels is more accurate than the Gamma-Gamma modeled channels because the coefficients of the Gamma-Gamma model tend to change more rapidly in time than the coefficients of the Log-normal model.