Scaling of input-output data and number of conditionality of a matrix of Grama

Is considered procedure of scaling of the input and output data for increase of accuracy of estimations in problems of parametrical identification of the objects described by models in the form of space state. Offered procedure allows to eliminate influence of property of small changes of sequence of measurements concerning some constant. Results of calculations of numbers of conditionality of information matrixes Grama received with the help a component of a vector of input and output variables without scaling and by data with preliminary procedure of scaling are resulted.