Statistics, Coverage, and Improvement of Modelling via ANN of Radio Mobile Signal in Vegetated Channel in the 700-4000 MHz Band

In communication system design, predicting radio signal coverage in channels with vegetation is challenging due to the way vegetation absorbs, reflects, spreads, and depolarizes the signal. Empirical models have been developed to predict signal coverage in these channels, such as parks and urban squares with vegetation, which are common in cities. To study these scenarios, this paper presents narrowband measurements taken in a public square with varied vegetation, using five carrier frequencies in the 700-4000 MHz band, including those used for cellular services, Wi-Fi 6 (Wireless Fidelity), and 5G (Fifth Generation) networks. Two transmitter antenna heights were used, and the measured data were analyzed graphically and quantitatively. The study found that an artificial neural network improved the signal prediction beyond that of the log-distance model, which is the basic model with the best results. Therefore, it can be used understand and model this kind of environment.