You can follow the classic Box-Jenkins ARMA approach using difference stationary time series. I also agree with Singh that difference stationary time series has a linear trend, which is why it is so easy to get it into stationary position unlike those with a stochastic trend. There is also a special case of difference called seasonal difference used for seasonality. In comparison, trend stationary requires you to estimate first then remove the trend from time series to get it into stationary.