On the background subtraction in 'low-dose' measurements using thermoluminescence dosemeters

The role and variance of various background subtraction methods have been discussed and examples given on the basis of various measurements performed on LiF pellets. For low-dose measurements the best method is that with smallest standard deviation. It is considered that for low dose measurements the background should be determined from a subpopulation of dosemeters of the same reading cycle. A further considerable decrease in the variance of the background can be achieved if both the unirradiated subpopulation and the irradiated dosemeters are read out twice, and the difference of the two readings of the same dosemeter used for further calculation. It is pointed out that the second reading of an unirradiated dosemeters is always less than the first one. This result may offer, in same cases, a reliable background subtraction method with small variance. (author)