Statistical Data Processing in Rocket-Space Technology

Abstract

The most measured data in the rocket-and-space technics is a broadband non-stationary random process. Using traditional methods of cyclic digitalization at processing of such data leads to heavy computational costs and heavy amounts of memory of on-board computing devices. As a rule, in processing of discontinuous processes, the recovery of original implementation on the receiving side is not required and processing involves calculation of probabilistic characteristics. In this case processing of data in the rocket-and-space technics has a number of features. First, random processes always represented by the sole implementation. Secondly, priori knowledge of measured random process probabilistic properties is not possible. Thirdly, there is need in on-line processing that predetermines use of rapid processing methods. The article considers application possibilities of nonparametric solution theory methods for evaluation of probabilistic properties of non-stationary broadband random processes in the rocket-and-space technics on-board systems.