The application of B-spline wavelet multi-resolution analysis (MRA) in UV spectrophotometry is presented, with a modified approach to signal processing. The results indicate that if the optimum B-spline wavelet and truncation frequency are selected, the absolute values of the peak relative errors are less than 2% when the signal-to-noise ratio is not less than 0.5, which means that useful signals can be separated from high-frequency noise. Most often third-order B-spline wavelet MRA is the most effective, In this respect, significant differences are observed for the different B-spline wavelets, The processed results of experiment data are also satisfactory.
