Prediction of Radio Wave Attenuation due to Cloud Using Machine Learning Techniques

The latest development in wireless technology has resulted in a surge in demand for higher frequency bands from all corners of the mobile industry. As next-generation mobile technology advance at a breakneck pace and the world moves to an online platform, technologies that provide faster internet with no lag are needed. Owing to the availability of higher bandwidth, millimetre waves and sub-millimeter waves are better candidates for this form of operation. These higher frequencies are hampered by environmental attenuation caused by rain, fog, dust, and other factors. In the case of satellite communication, cloud-induced radio wave attenuation is important. For calculating attenuation, various models such as ITU-R, Slobin, Gunn, and others are available, but ITU-R is the most commonly accepted. Water droplet dielectric constants are determined by calculating attenuation using the ITU-R model. Using machine learning techniques, a new method for measuring the real and imaginary parts of the dielectric constant of a water droplet is presented in this paper. The proposed model's results are compared to the ITU-R model's. In comparison to the ITU-R model, the proposed model has the advantage of being very straightforward since it includes quadric equations.