Satellite-UMTS traffic dimensioning and resource management technique analysis

Abstract

The universal mobile telecommunications system (UMTS) will consist of space UMTS (S-UMTS) and terrestrial UMTS (T-UMTS) components. An algorithm for predicting the traffic capacity in terms of the number of subscribers for the satellite component of UMTS is presented. The algorithm takes into account the takeup characteristics of new products, the growth of gross domestic product (GDP), the projection of population, the tariff of the service, and price fall over the forecast period. The predicted traffic is used to generate a traffic grid in terms of Erlang of dimension 36/spl times/72 in steps of 5/spl deg/ in both the latitude and longitude directions. The traffic grid is used to evaluate the performance of a dynamic channel allocation (DCA) technique as, well as a fixed channel allocation (FCA) technique. Both channel allocation techniques have been considered with the queuing of handover (QH) requests. In order to compare the respective techniques' performance, a low-earth orbit mobile satellite system (LEO-MSS) mobility model is developed to take into account the effect of satellites' motion during interbeam handovers. A theoretical model for obtaining the values of blocking probabilities for low-traffic loads is presented. Finally, the performance of the DCA-QH technique is compared with the FCA-QH technique under suitably defined traffic and mobility conditions.