FLEXIBLE RESOURCE ALLOCATION TECHNIQUES FOR NGSO CONSTELLATIONS (ARTES AT 3A.095)

Status

ISSUED

EMITS REFERENCE

Program

ARTES Advanced Technology

Price Range

> 500 KEURO

Description

 

Objective:

The objectives of the activity are to develop a resource management algorithm able to optimise the allocation of resources of a non-Geostationary (NGSO) satellite constellation based on the geographical distribution and variation in time of the traffic demand and to demonstrate that such algorithm can converge fast enough so that it would be operationally useful.

A software system demonstrator implementing the resource management algorithm will be developed and tested.

Targeted Improvements:

To increase the system throughput 2-3 fold by allocating the available resources where and when needed contrary to the classic uniform resource allocation that ignores the geographical distribution and variation in time of the traffic demand.

Description:

In the last years, a number of NGSO constellations for providing telecommunication services have been announced. As it has been experienced in existing satcom systems the traffic demand is not uniformly distributed. Therefore the actual system throughput can be severely limited. Satellite networks need to embark flexible payloads allowing them to adapt their offer of capacity to the actual distribution of the traffic demand and thus maximise their throughput. In other words, they should be able to offer capacity where needed and when needed. Constellations are also subject to this fact.

Adapting the offer of capacity to a given distribution of the traffic demand not only requires flexible payloads that allow the dynamic allocation of resources, but also to find the optimal allocation of such resources. Furthermore, in an operational environment it requires an algorithm capable of finding an allocation that is “good enough” in a reasonable amount of time. Several approaches to solve the problem are possible. Heuristic strategies are the simplest and would allow to converge to a suboptimal solution in a single iteration. However, more advanced iterative optimisation algorithms would allow reaching better performance at the cost of a higher complexity and longer convergence times.

This activity will develop resource management and allocation algorithms for NGSO constellations. Such algorithms will be a balance between system performance (with respect to the optimal solution) and convergence time. The activity will implement those algorithms and assess their performance in a software system demonstrator.

Tender Specifics