Propagation of solar activity using the block approximation method, generalized weighed root-mean-square methods and revealing latent periodicities methods

Abstract

The low Earth orbit propagation accuracy critically depends on the knowledge of atmosphere density variations determining the atmosphere drag to the low Earth satellites. In its turn, the atmosphere drag depends on the intensity of solar radiation and is the most important factor of determining the orbiting lifetime of a satellite in the low Earth orbit or the geostationary transfer orbit (having a very low perigee). However, the prediction of the atmosphere density variations is a weak point of the low satellite motion parameters propagation. The processes of such a kind are characterized by instability of the registered signal level, by presence of appreciable stochastic component, and often by absence of a priori knowledge on their analytic structure. Application of classic methods in such cases usually does not give satisfactory results. In this paper, a family of methods for enhancing the propagation of a wide class of processes having non-stable parameters is proposed. The methods are of a phenomenological type (not requiring any a priori information) including the block approximation method, D-methods of revealing latent periodicities, the generalized weighed root-mean-square method. These methods allow revealing the real structure of processes investigated by their measurements only. Moreover, they give a possibility for concretizing obscure and approximate assumption on the process’s structure, as well as enhancing the quality of prediction of processes having roughly determined or “floating” values of their parameters and to guess the trend of their variations. As an example of application of the proposed methods the solar flux projection is made with their help.