INDEPENDENT COMPONENT ANALYSIS AND ITS APPLICATION FOR PREPROCESSING NDS
Jianping Wu · Computing Techniques for Geophysical and Geochemical Exploration · 2008
Nuclear detector signal processing has all along been a hard task because of its poor signal to noise ratio (SNR) of the raw nuclear detector signals and almost about the latent targets what ones want to detect beforehand. A new approach that uses Independent Component Analysis(ICA), a newly developing blind signal processing(BSP)technique, to nuclear detector signal processing has been proposed in this paper。With the help of a fast ICA algorithm, it has successfully extracted quite weak target signals from the nuclear detector signal data with much strong background. So that the SNR of the nuclear detector signal and the detection performance of the nuclear detector signal have both improved greatly. The ambiguities of the sign uncertainty problems of separated target signal components in ICA practical use have also been overcome. The works show the good perspective of ICA application on nuclear detector signal processing.