Algorithm of radionuclide activities estimation for 
scintillation spectra with taking into 
account sample's density 

S.A. Tolstov, S.V. Linev, V.A. Muravsky

     It is  known   that   a   variation   of   the   density   of   a
space-distributed  radioactive sample (source) influences the shape of
the nuclides radiation spectrum.  An algorithm for nuclides activities
estimation has been developed,  which takes into account variations of
sample density.  This algorithm is based  on  the  maximum  likelihood
method and includes a linear transformation between the vectors of the
activities estimates for different densities.  The algorithm developed
does  not  depend  on  the  type  of  ionizing  radiation and requires
substantially  less  volume  of  calibration  data  to  be  stored  in
comparison  with  the standard methods of taking into account a sample
density.  It needs only initial spectra of the nuclides radiation  for
single  magnitude  of  sample  density and a set of the transformation
matrixes for series of densities to be stored.