Inversion of PM2.5 atmospheric refractivity profile based on AlexNet model from the perspective of electromagnetic wave propagation

Guo, ChengJun; Xu, YaLan; Tian, Zhong

Inversion of PM2.5 atmospheric refractivity profile based on AlexNet model from the perspective of electromagnetic wave propagation

2020

Guo, ChengJun | Xu, YaLan | Tian, Zhong

Human civilization has reached an unprecedented height, but the industrialization of economic development also brings global warming, ozone depletion, acid rain, fresh water resources crisis, energy shortage, and environmental problems. In autumn and winter, haze becomes the usual state in the modern society, and PM2.5 has been becoming an important form of air pollution. The research found that PM2.5 brings great influence to the human body or daily life. To some extent, the PM2.5 also affects the propagation of electromagnetic waves near the ground, reducing the transmission performance of electromagnetic wave. Based on Mie scattering theory, this paper qualitatively analyzed the scattering effects of PM2.5 particles on every frequency band of electromagnetic wave in daily use. Then the paper takes the satellite navigation signals as a research example, selecting university of Wyoming Davis stations in Antarctica sounding data by measuring the tropospheric atmospheric meteorological parameters (including the atmosphere pressure, geopotential height of different layers, dew point temperature, relative humidity and specific humidity, wind direction, wind speed, and temperature). The paper inversed the refractive index distribution of the troposphere based on AlexNet model and described the error quantitatively. The simulation results show that the estimated error is less than 5.1455%, proving the high accuracy of the AlexNet model. To test the influence of PM2.5, the paper takes Jiuquan, a city with serious pollution, as an example. Comparison between the inversion results and IGS products shows that high concentration of PM2.5 pollution has little influence on the inversion of refractive index profile.

Mostrar más [+]