Abstract: The correlation coefficient stationary process in which the mean and variance vary with time is familiar in engineering,and the traditional correlation function stationary process is just a special case of that.The method of time frequency analysis and AR(p),MA(q) and ARMA(p, q) models of the correlation coefficient stationary process are presented in this paper.Its mean function, variance function and correlation coefficient function can be gained by the method in time domain.Thus the spectrum density of the correlation coefficient stationary process can be obtained by Fourier transform,short time Fourier transform or wavelet transform.The formulas for calculating the response of a linear system are derived when the input is a correlation coefficient stationary signal.In addition, the concepts of trend component spectrum density and random component spectrum density in random processes are also presented herein.The method has been used in the performance tests of aircraft and satellite successfully.It shows that a great number of specimens and time have been saved.