Data smoothing using a least squares fit C++ class

Abstract Two types of data smoothing algorithms are presented which reduce unwanted noise from raw data collected by means of a data acquisition system. One algorithm (moving average) averages the magnitude of a specific data point with a preset number of neighboring data points. A second algorithm presented employs a least squares fit criterion to smooth data. The least squares fit algorithm presented employs tables of convolution integers and normalization factors to calculate smoothed data points. A brief introduction to C++ class structure is included along with source listing of a data smoothing class.