Statistical characteristics and basic model of atmospheric turbulence are reported Using random number sequence generated by randomizer of SVS-Fortran as random variable, random process of atmospheric turbulence is objectively described A equivalent model of horizontal atmospheric turbulence is constructed through FFT and filtering. Basing on Karman spectra model of atmospheric index-of-refraction power spectra density and the Generalized Huygens-FYesnel diffraction integral theory, the process of laser propagation in atmospheric turbulence is simulated with an efficient two-dimensional FFT and IFFT by computer. It is shown that additory phase generated by random phase screens reflected characteristic parameter of structure of atmospheric turbulence.