An automatic recognition approach of constellation satellites' usability based on clustering

Abstract

Satellites' usability may change with time due to collision, long time service, and so on. Thus, the target satellites should be analyzed frequently in order to monitor their health. Though the approach based on detecting satellite signals is the most reliable and widely applied, it has the disadvantages of heavy workload, long period, low efficiency and poor real-time capability. To solve this problem, this paper presents an approach to recognize constellation satellites' usability automatically by computer using clustering. According to the relationship between satellite's usability and its orbit characteristics, using the latest orbit parameters downloaded from the internet, this approach recognizes the satellites according to their differences in orbit characteristics. This approach has high speed and little manual intervention, which can be used to all the constellations that follow a certain orbit rule, such as Iridium, Globalstar, GPS and so on. Take Iridium for example, the correct rate is 80.44%.