Abstract: The first prerequisite for obtaining correct information from an X-ray diffraction pattern is to accurately remove the background. The most commonly used method in experiments is direct subtraction, which ignores the absorption of the sample and other interactions within the sample such as the photoelectric effect and the counting of photons produced by Compton scattering. The "rolling ball" algorithm and the use of software such as Jade deduction background as well as other deduction methods also have their own limitations. Monte Carlo simulation can identify and count the particles, so it is very important to use it for quantitative analysis of XRD background. In this study, the atomic Form Factor of each element in the air was calculated by the Baró fitting formula and the molecular Form Factor was given by the Debye formula and compared with the calculation results of the independent atomic model. Geant4 was used to simulate XRD for crystalline substances such as bone and nylon and amorphous substances such as water, and the background and real signals were effectively separated in the results, then the XRD background was quantitatively analyzed on this basis. The results show that there is a huge difference in counting between the independent atomic model and Debye's formula,and the Debye method produces more scattered particles. XRD background comes from the air scattering and other types of interactions between photons and matter in the sample, the former contributing the most at small angles, the proportion decreases from 100% to less than 40% with the increase of the scattering angle, and the proportion of the latter increases from 0 to more than 60% with the scattering angle. The results indirectly illustrate the inaccuracy of the background subtraction methods in the experiment. Such as the direct subtraction method not only ignores the absorption of X-ray by the sample, but also ignores the background count caused by other interactions of X-ray photons in the sample. To accurately remove the background, all the counts of background mentioned above must be considered. The Monte Carlo method was used to simulate the XRD background, and the simulation was closely connected with the practice. The same XRD background pattern can be obtained by constructing the same geometric model as the experiment, setting the same physical parameters as the experiment, and using the same sample Form Factor as the experiment. Applying the simulation results to actual experiments may become a more accurate method of background deduction.