A novel estimation of MMW sky brightness temperature based on BP neural network

In millimeter-wave (MMW) passive detection, the sky brightness temperature is a crucial physical quantity that usually determines parameters of the detecting system. It presents great nonlinear relations with frequency, zenith angle and meteorological conditions. In this paper, a novel method utilizing the nonlinear approximation function of the back propagation (BP) neural network was proposed to estimate the sky brightness temperature at MMW band. The given simulation results demonstrate this method can obtain a higher degree of accuracy than the previous methods, especially for the large zenith angle conditions. Furthermore, the proposed method avoids the complicated calculation of the atmospheric absorbing coefficient and provides an innovative way to estimate the MMW sky brightness temperature.