In 2008, Stephen put forward a new brightness temperature function and the expression is purely empirical. However, it does not require knowledge of the total amount of atmospheric water vapor in retrieval of cloud-base height. Systematical investigation of the changes of the cloud-based infrared brightness temperature Tb as the zenith angle, water vapor, cloud, aerosol, visibility and other meteorological conditions is performed by using the SBDART model and Stephen brightness temperature function, and the error is analyzed. The results show that zenith angle, water vapor, cloud, aerosol and visibility less than 5 km have a very important influence on Tb. Under different meteorological conditions, the simulation results of Stephen function are similar to SBDART model. Therefore, Stephen brightness temperature function can be used to simulate the cloud-base brightness temperature.