Naslov
A Heuristic Approach to Satellite Link Capacity Planning Applied in Mobile Networks (Heuristički pristup u planiranju kapaciteta satelitskog linka mobilnih mreža)
(A Heuristic Approach to Satellite Link Capacity Planning Applied in Mobile Networks)

Autori
Krile, Srećko

Izvornik
Promet-Traffic-Traffico (0353-5320) 15 (2003), 1; 19-29

Vrsta, podvrsta i kategorija rada
Radovi u časopisima, prethodno priopćenje, znanstveni

Ključne riječi
optimal capacity expansion; satellite link capacity planning; mobile satellite networks; QoS in mobile satellite networks.
(optimal capacity expansion; satellite link capacity planning; mobile satellite networks; QoS in mobile satellite networks)

Sažetak
The efficient heuristic algorithm for sizing of N satellite capacity types (for LES - satellite link) in mobile networks during exploitation is being developed, minimizing the total expansion cost. Using the network flow approach and the extreme flow theory many expansion solutions (sub-problems) are eliminated from further calculation, because they consist at least of one flow that cannot be a part of optimal expansion sequence. At first the heuristic algorithm is compared with algorithm based on exact approach. In all numerical test-examples the best possible result is achieved. After that we developed and tested some algorithm options, using various limitations for capacity state values. It is obvious that all heuristic options are very effective and for some cases they are capable to find the best possible result but with significant computation time savings. It means that this heuristic approach can be successfully applied to short-term or medium-term satellite network planning with finite number of discrete time periods. Only If we apply adequate heuristic solution we can ensure both, significant improvement of QoS (Quality of Service) for capacity availability and minimal capacity expansion cost.

Izvorni jezik
Engleski

Znanstvena područja
Elektrotehnika, Računarstvo, Tehnologija prometa i transport

POVEZANOST RADA