An efficient k–level traffic control scheme with congestion awareness in wireless mesh networks. (4th March 2013)

• Record Type: Journal Article
• Authors: Dow, C.R.; Lin, P.J.; Hwang, S.F.
• Is Part Of: International journal of ad hoc and ubiquitous computing. Volume 12:Number 3(2013)
• Page Start: 129
• Page End: 140
• Publication Date: 2013-03-04
• Subjects: WMNs -- k–level congestion awareness -- fuzzy control -- MMPP -- Markov–modulated Poisson process -- k–level traffic control -- wireless mesh networks -- wireless networks -- fuzzy logic -- traffic burstiness -- simulation; Ubiquitous computing -- Periodicals; Embedded computer systems -- Periodicals; Electronic data processing -- Distributed processing -- Periodicals; Wireless communication systems -- Periodicals; Computer architecture -- Periodicals; 004.2
• Languages: English
• ISSNs: 1743-8225

Abstract

Wireless mesh networks have attracted significant attention due to their ability to achieve a wide–area wireless environment with flexible configuration and low–cost properties. With utilisation of Wireless Mesh Networks (WMNs) as an infrastructure for internet access, all network traffic from mesh routers are forwarded to gateways directly connected to the wired network. Communication in large–scale wireless networks may suffer capacity bottlenecks in the gateway and local congestions among mesh routers. This paper addresses the issue by presenting a hybrid strategy that combines the gateway–based and path–based traffic control schemes. To obtain simple and efficient results, the proposed k–level congestion awareness strategy uses the fuzzy logic technique to estimate gateway loading. Border nodes redirect partial traffic to adjacent regions belonging to neighbouring gateways when congestion occurs. The proposed scheme can adjust the network traffic automatically according to traffic conditions, for smooth and immediate sharing of the traffic load to neighbouring gateways, making the k–level congestion awareness strategy an adaptive approach for solving gateway bottlenecks and local congestion issues. This study also uses a well–known Markov–Modulated Poisson process (MMPP) to capture traffic burstiness and correlation. Simulation experiments have ascertained the model's validity and results have indicated that the proposed scheme is a sound option for WMNs to support congestion control, especially in bursty traffic environments.