Scarceness of bandwidth is a common problem to all radio networks. The cellular structure with frequency reuse has been the solution for many years in the past; however, decreasing cell sizes results in increasing handoff requests and this could be critical in high mobility environments. Moreover, the cellular structure cannot ensure the best utilization of resources when the distribution of the users inside the system is not constant in time. In this paper we focus on handoff protection and efficient dynamic bandwidth allocation for multi-service networks in high-mobility environments, in presence of very simple strategies for the Call Admission Control, in order to assure scalability. We propose to find an optimal solution by using a new simple heuristic which exploits the prediction of the system behaviour made by some suitable network model; moreover, a more traditional approach based on a stochastic algorithm is derived from the Montecarlo methods for comparison. Simulation results show that the proposed heuristic approach performs better, especially in critical situations.
Dynamic Bandwidth Allocation for Wireless Networks in High-Mobility Environments
BOLLA, RAFFAELE;REPETTO, MATTEO
2006-01-01
Abstract
Scarceness of bandwidth is a common problem to all radio networks. The cellular structure with frequency reuse has been the solution for many years in the past; however, decreasing cell sizes results in increasing handoff requests and this could be critical in high mobility environments. Moreover, the cellular structure cannot ensure the best utilization of resources when the distribution of the users inside the system is not constant in time. In this paper we focus on handoff protection and efficient dynamic bandwidth allocation for multi-service networks in high-mobility environments, in presence of very simple strategies for the Call Admission Control, in order to assure scalability. We propose to find an optimal solution by using a new simple heuristic which exploits the prediction of the system behaviour made by some suitable network model; moreover, a more traditional approach based on a stochastic algorithm is derived from the Montecarlo methods for comparison. Simulation results show that the proposed heuristic approach performs better, especially in critical situations.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.