GaliLEO. A simulation tool for traffic on LEO satellite constellations. Presentation of the architecture

Celandroni N., Ferro E., Potortì F., Franck L.

COST  LEO  Satellite  Simulation 

Complex communication systems where satellite links are involved are difficult to test and tune up without the help of simulation tools. First of all, using satellites is very expensive - the satellite time spent in testing and tuning-up the system must thus be as short as possible. Second, during the performance evaluation in a real environment it is not always possible to find the right amount of traffic, and the most appropriate traffic pattern and data aggregation that will put the system under the maximum amount of stress so that its limits can be validated. When satellite constellations are involved, rather than single satellites, these problems are enormously exacerbated, to the point that in all but very particular cases it is practically impossible for the researcher to do tests on satellite constellations. It is therefore necessary to use some kind of simulation tool. The array of required features for the tool is very broad. The basic feature is that of making it possible to describe the constellation topology, and the position of earth terminals. Satellite may have different capabilities, in term of number and characteristics of intersatellite links, spot topology and behaviour, and on-board processing abilities. Many types of routing algorithms are under study, some of which allow for dynamic rerouting based on nodes' perceived congestion or on quality of service requests. Also, it should be possible to analyse channel allocation algorithms for the up-down links, including frequency reuse algorithms. The traffic generation choice should be broad and easily extensible, allowing for both fine control on single connections and generation of background traffic for loading the satellite network. Finally, a rich set of statistic collection and analysis tools should be available.