Key points are not available for this paper at this time.
An attempt is made to give a coherent, elementary account of the ways in which fluctuation theory has been applied to some of the simpler types of physical measurement. Uncertainties in measurements involving suspended systems are discussed on the basis of simple correlation function arguments, and consideration is given to the methods of measurement appropriate when there are various practical limitations on the parameters of the suspended system. There follows an elementary development of the mathematical and physical considerations usually employed in treating fluctuations in linear measuring instruments: remarks on the status of the random force method as applied to equilibrium systems are included. The circumstances under which there is an absolute limit to the attainable accuracy are discussed. It is shown that, in the case of measurements with a suspended system, feedback may enable these limits to be attained with a convenient measurement procedure. An account is given of various approaches to the calculation of the limits of accuracy in measurements with radiation detectors. Finally some elementary results are established concerning the optimum characteristics of an instrument used to follow a varying signal in the presence of noise.
C. McCombie (Thu,) studied this question.