Spline wavelet self-convolution in processing overlapped peaks in capillary electrophoresis
Abstract
The normal distribution function was used as the resolving factor to act on the filters of the spline wavelet to produce the peak resoluter of spline wavelet self-convolution (SWSC). Several types of simulated overlapped peaks were processed by this peak resoluter and satisfactory results were obtained. Baseline separation can be achieved and overlapped peaks can be separated directly in the time domain. The relative errors of the peak positions and areas between the original peaks and the processed peaks are less than 5.0%. The overlapped peaks of capillary electrophoresis signals for amino acid mixed solutions and inorganic ion solutions were resolved by this method and satisfactory results were also obtained.