Analysing the correlation performance of ESA's planned space-based GEO surveillance mission

The surveillance of the region around the geosynchronous orbit (GSO) is usually done with optical sensors, i.e. telescopes. Placing a telescope on a satellite platform and performing the observations from space has several advantages, such as independence from weather and possibly an extended coverage. Several operational sensors of this type already exist, e.g. Sapphire and SBV. The European Space Agency is also supporting the development of technologies for such a mission. The envisaged mission design would be to place the sensor on a sun-synchronous orbit close to the terminator plane to observe objects in GSO passing through dedicated fences in the anti-solar direction. In this work the correlation of tracklets is analysed, which means to test whether two tracklets originate from the same object. This is done by iterating on the hypothetical ranges and solving a boundary value problem. The results show that an initial catalogue build-up with observations only from such a sensor would be problematic because a large number of false associations is made. Even a post-processing technique based on the search for clusters in the correlations cannot reduce these false associations sufficiently. In total approx. 70% of the objects can be identified as unique clusters. Other possible influences which are not investigated in detail is the seasonal dependence of the observations and the sensitivity of the correlation to higher noise levels of the sensor. It is concluded that a sensor used with the given mission design would probably be more useful for catalogue maintenance than for catalogue build-up. However, this also has to be seen in the framework of a complete surveillance network.