Abstract
The determination of electron temperature and density profiles with high resolution Thomson Scattering is reconsidered, focusing on the correct treatment of the statistics and error estimation. Based on this, methods are proposed to decide whether spatial structures observed in the measured profiles are significant. A detailed statistical analysis of the detection process is given. It is shown that a maximum likelihood method must be used to fit the theoretical scattering spectra to the data, in order to obtain electron temperature and density. Furthermore, an existing calibration procedure for the detector, based exclusively on the statistics of the detected scattering image, is perfected. This self-consistent analysis leads to a precise definition of the confidence limits. Based on these, statistical tests are designed to determine whether structures seen on the Thomson scattering profiles are significant or not. To test the statistical method, some TEXTOR discharges were analyzed.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
