Precising the evaluation of radio information system reliability characteristics based on the empirical failure distribution

Abstract

For highly reliable radio information systems the failures are quite rare phenomena leading to critical decrease in functional characteristics. At the same time, failure distribution may be a subject to rather complex laws, which together with a lack of real data makes it difficult to evaluate the reliability characteristics of the product. Achieving high accuracy of reliability characteristics evaluation is possible due to constant updating the parameters of the radio information system reliability model based on experimental data on the technical condition of the product components starting from the stage of its development. In this case, it is possible to determine the reliability characteristics based on current data on the performance of radio information system components based on the empirical failure distribution function. In this article we propose an algorithm for transferring to an empirical distribution and counting the characteristics (mean time between failures) of radio information system reliability based on it.