Design and Analysis of Mobility-Aware Clustering Algorithms for Wireless Mesh Networks

Nizar Bouabdallah (1) , Rami Langar (2) , Raouf Boutaba (3)

1 DIONYSOS - Dependability Interoperability and perfOrmance aNalYsiS Of networkS
2 Phare
3 University of Waterloo [Waterloo]

One of the major concerns in wireless mesh networks (WMNs) is the radio resource utilization efficiency, which can be enhanced by efficiently managing the mobility of users. To achieve this, we propose in this paper the use of mobility-aware clustering. The main idea behind WMN clustering is to restrict a major part of the exchanged signaling messages due to the mobility of users to a local area (i.e., inside a cluster). As such, less wireless links are used by the signaling messages, which reduces the resources occupied by a mobile user during its service and thus improves the total network capacity. Through analytical models and simulations, this work first identifies the cases where clustering is helpful. Building on these results, we propose two clustering schemes that take into consideration the mobility properties of the users in order to improve the WMN performance. We prove that both schemes can achieve significant gains in terms of radio resource utilization. Specifically, we show that the first scheme fits better both large and low-connected wireless mesh networks, whereas the second scheme is more suitable for both small to moderate and highly connected networks.

Domains: Networking and Internet Architecture [cs.NI]

Nizar Bouabdallah, Rami Langar, Raouf Boutaba. Design and Analysis of Mobility-Aware Clustering Algorithms for Wireless Mesh Networks. IEEE/ACM Transactions on Networking, 2010, 18 (6), pp.1677-1690. ⟨10.1109/TNET.2010.2049579⟩. ⟨hal-00793307⟩