D.S. Andrashitov - Ph.D (Eng.), Academy of Strategic Rocket Forces named after Peter the Great. E-mail: Dima-andrahitov@rambler.ru
I.V. Deryabkin - Post-graduate Student, The South-Russian State University of Economics and Service. E-mail: i.deryabkin@jint.biz
А.А. Kostoglotov - Dr.Sc. (Eng.), Professor, The South-Russian State University of Economics and Service. E-mail: kostoglotov@aaanet.ru
S.V. Lazarenko - Ph.D (Eng.), Senior Lecturer, The South-Russian State University of Economics and Service. E-mail: rh3311@mail.ru
B.М. Tsennykh - Post-graduate Student, Rostov State Transport University. E-mail: rh2010@mail.ru

Due to the rapid development of ADP equipment the task of optimization of metrological characteristics of means of measuring tech¬nique is connected to optimum processing of measuring information. Such measuring tasks from the point of view of causes and effect relationships are reverse and are brought together to estimation of some function by results of, as a rule, dynamic measure¬ments. At the heart of setting of the reverse measuring task the choice of the mathematical model providing the quantitative description of researched process lies. In article the new method of the solution of the reverse measuring task is offered. In its basis the assumption lies that dynamics of all proceeding processes satisfies to the equations of Lagrange of the second kind. Need of the accounting of restrictions following from it requires use of a method of multipliers of Lagrange that allows to construct an expanded target functionality. Results of the solution of such received extremum problem allow to receive structure of adjusting back coupling of the accelerometer in the closed form. Its use provides receiving estimates of parameters of the accelerometer meeting to the real values and an input measuring signal that is confirmed by results of mathematical simulation. The scoring in volume of computing expenses in comparison with Kalman's filter is thus provided.