Solar radio storm intensity judgment method based on SVM multi-classification algorithm

Abstract

The invention discloses a solar radio storm intensity judgment method based on an SVM multi-classification algorithm. The basic principle of the SVM multi-classification algorithm is that for a multi-classification problem, every two types are combined into a binary classifier, for each binary classifier, an optimal hyperplane is found out, samples are divided into two types, and finally a classification result is obtained through a voting statistics method. In the process, firstly, data capable of reflecting the solar radio storm intensity of an observed earth satellite are extracted, data preprocessing is carried out, a feature vector is obtained, and the solar radio storm intensity type is marked. And the samples are input into an SVM classifier for learning to obtain an optimal classifier. And when the new feature vector enters the classifier, automatic classification is carried out. The determination method can automatically determine the solar radio storm intensity type, is highin efficiency and accuracy, does not depend on a radio telescope, and is low in cost.

Description

technical field [0001] The invention relates to the technical field of wireless communication, in particular to a method for judging the intensity of solar radio storms based on an SVM multi-classification algorithm. Background technique [0002] With the increasing application of GNSS and other satellite technologies in modern society, the influence of solar radio storm intensity on GNSS signals has become an important part that cannot be ignored. Since radio bursts usually occur very suddenly, and often occur at the same time as proton bursts, X-ray bursts, or flare bursts in the solar active region, the radiation intensity is large and changes drastically. When it reaches the earth, it will cause a series of earth Physical effects, such as magnetic storms, lasers, communication interference, etc. During many solar radio bursts in the past, the solar radio burst may lead to a decrease in the satellite carrier-to-noise ratio, an increase in positioning error, an increase i...

AI Technical Summary

Problems solved by technology

[0004] Traditional classification methods generally require astronomers with professional knowledge to manually classif

Images