Uncertainty analysis of the inverse LASSO estimation scheme on radioactive source distributions inside reactor building rooms from air does rate measurements

Abstract

Very recently, Least Absolute Shrinkage and Selection Operator (LASSO) has been proposed as a scheme capable to inversely estimate radioactive source distributions inside reactor building rooms from air dose rate measurements together with the predicted lower bound of the measurement numbers for successful reconstructions. However, no one has ever analyzed how the uncertainty of input data including the measurement errors influences the accuracy of the inverse estimation results. In this paper, we therefore perform uncertainty analysis of the LASSO scheme and suggest an uncertainty estimation function derived based on the theory of Candes. We actually demonstrate in two types of numerical tests with different input uncertainties obtained by using Monte Carlo code, Particle and Heavy Ion Transport code System (PHITS) that the calculated errors obey the proposed uncertainty estimation function. Thus, the LASSO scheme allows to successfully estimate radioactive distributions within the predicted uncertainty.