A retrieval of the drop size distribution from microwave link signal fluctuations

Abstract

Mobile phone, and other communication networks transmit information using microwave propagation. Recent research has shown that these microwave links can be used to measure rainfall intensity by using the signal attenuation between a transmitting and receiving antenna. This method has great potential due to the abundance of mobile phone networks all over the world. A known issue with this method is that wet antenna due to rainfall often causes an overestimation of the rainfall intensity. We present a method of rainfall retrieval from these links that is in principle insensitive to wet antenna, and could ultimately lead to much improved rainfall retrievals. Instead of looking at the attenuation of a signal, we propose to look at the fluctuations of the signal caused by raindrops. We use the fact that rain falls through the first Fresnel zone of the microwave link, which causes signal fluctuations. We derive expressions for signal fluctuations as a function of frequency, path length and raindrop size distribution. We can then compare theoretical predictions with microwave link data. Using statistical inference methods, estimates for the drop size distribution can then be retrieved, and from this the rainfall intensity.