A.N. Dolgov, V.N. Churkin, O.V. Shvydya

The main feature of modern monitoring systems composition is their complexity, distribution and large resource intensity. The projecting of such systems, as a rule, is carried out within the scope of aggregative-decomposition analysis. The article deals with the approach to the substantiation of decomposition principles and step-by-step projecting of distributed monitoring systems. It is demonstrated, that the solution of a set problem is based upon the research of additivity properties of general-system and particular quality measures of their functioning. Within the framework of Bayes risk the assessment model of monitoring systems effectiveness is proposed. The system decomposition feasibility vertically and on the level as well as independent project analysis of its composite elements are demonstrated in terms of general-system results: carrying capacity, system cost, recognition probability of independent state parameters of controlled means and technological processing as well as keeping information value probability for the time being in the system. This possibility allows to carry out staged and bit-by-bit organizational-technological system improvement taking into account priority project tasking and resources restrictions. It is demonstrated, that the decentralized data and information processing in the monitoring system is more effective than the centralized one, for equal recognition probability it can considerably reduce the requirements to the carrying capacity (cost) of the signal channels and increase the monitoring system survivability.