Ionosphere scintillation in GNSS signal can be characterized by scintillation indices computed from the power and phase measurements obtained from the receiver tracking loop outputs. The effectiveness of the scintillation indices is a function of the detrending method used to separate the slowly varying receiver clock, satellite dynamics and low-frequency multipath from the higher frequency scintillations present in the measurements. The prevalent detrending method is a 6 order Butterworth filter with a 0.1 Hz cut-off frequency. The effectiveness of this method has been questioned by a number of researchers. In this paper, a discrete wavelet transform based method is developed as an alternative means for detrending. By applying the two methods on simulated GPS signals, we found that the Butterworth filtering tends to overestimate the scintillation indices while the wavelet filtering tends to underestimate them. Additionally, the wavelet method generates results that are closer to the actual indices’ values. Both methods are applied to real GPS scintillation data collected in Alaska. Furthermore, the wavelet filtering leads to a larger correlation between amplitude and phase scintillation.