Statistical modeling of deconvolution procedures for improving the resolution of measuring electron temperature profiles in tokamak plasmas by Thomson scattering lidar

摘要

The potentialities are investigated, by statistical modeling, of deconvolution techniques for high-resolution restoration of electron temperature profiles in fusion plasma reactors like Joint European Torus (JET) measured by Thomson scattering lidar using the center-of-mass wavelength approach. The sensing laser pulse shape and the receiving-system response function are assumed to be exponentially-shaped. The plasma light background influence is taken into account as well as the Poisson fluctuations of the photoelectron number after the photocathode enhanced in the process of cascade multiplying in the employed microchannel photomultiplier tube. It is shown that the Fourier-deconvolution of the measured long-pulse (lidar-response-convolved) lidar profiles, at relatively high and low signal-to-noise ratios, ensures a higher accuracy of recovering the electron temperature profiles with three times higher range resolution compared to the case without deconvolution. The final resolution scale is determined by the width of the window of an optimum monotone sharp-cutoff digital noise-suppressing (noise-controlling) filter applied to the measured lidar profiles.