A method for quantitative characterization of a phase's depth distribution is discussed in detail. Both model-independent and model-dependent nonlinear least squares technique methods were developed; in addition, an inverse Laplace transformation method is presented to solve the problem directly in mathematics. The methods can also be used for samples with preferred orientation. Furthermore, the technique is expanded to the technique of computed depth profiling of XRD patterns; then the depth profiles of other structural information that is based on the peak intensity, peak position, and line profile can be determined. A feasibility test was also performed.