About Optimal Algorithms for Making Statistical Decisions for Small Volume Samples and with Apriori Parametric Uncertainty

Development of systems of geoinformation monitoring demands the decision of some problems of the organisation of data flows of measurements. Among these problems of one of important the problem of acceptance of the statistical decision on presence on a surveyed part of a terrestrial surface of this or that phenomenon is. One of features of conditions of gathering of the information for such decision is the impossibility of reception statistical samples small volumes. Therefore working out and research of optimum algorithms of acceptance of statistical decisions for sample small volume are necessary at informational restrictions. For a case when the number of supervision is great enough, the problem dares a method of an estimation of parameters of likelihood distributions which is effective at unlimited growth of volumes sample on which basis the estimation of parameters is made. At the limited volumes sample, received by a method of an estimation of the parameters, the solving rule does not satisfy to necessary conditions of an optimality: To a constancy of average probability of an error of first kind and unbiasedness. In the present work the generalised adaptive algorithm of training to acceptance of statistical decisions for exponential groups of distributions is developed at aprioristic parametrical uncertainty of conditions samples small volume. Also estimates were obtained of the feasibility of the fixing block Remote Monitoring System (RMS) in the case of mobile anomalies.