Gateway-level Load Balancing Techniques for WMN: A Comparative Study Banani Das Department of IT Assam University Silchar, India banani.das.bd@gmail.com Amit Kumar Roy Department of IT Assam University Silchar, India amitkroy12@gmail.com Ajoy Kumar Khan Department of IT Assam University Silchar, India ajoyiitg@gmail.com Sudipta Roy Department of IT Assam University Silchar, India sudipta.it@gmail.com AbstractWireless mesh networks (WMNs) have emerged as a key technology for next-generation wireless networking, because of their advantages over other wireless networks. Due to the ever growing structure of WMN the traffic volume is expected to be very high, thus load balancing becomes a crucial part of it. Network load balancing enhances the scalability and availability of network. Load balancing can be used to extend the lifetime of a Mesh Network. Various approaches have been proposed by various researchers for load balancing between Internet Gateways (IGWs) and thus reducing the traffic congestion and improving the network performance and providing a better quality of service (QoS). This paper presents a detailed investigation on various Gateway level load balancing techniques for WMNs that have been proposed so far along with the critical comments as to which approach is better and suitable in which condition. Keywords— Wireless Mesh Networks (WMNs); Load Balancing; Internet Gateways (IGWs). I. INTRODUCTION Wireless mesh networking (WMNs) is the most advantageous next generation wireless networks. WMNs are undergoing rapid progress and inspiring numerous applications and have emerged as a key technology. WMN has been widely accepted as a replacement for areas of the MANET. WMN [1] is a communication network made up of radio nodes organized in a mesh topology and is a packet- switched network with a static wireless backbone. The topology of wireless backbone is fixed and modifications to infrastructure can only result from addition or removal or failure of access points. WMN consists of wireless access and wireless backbone network, in contrast to any other wireless networks. It is dynamically self-organized, self-configured, self-healing, easily maintainable, highly scalable and reliable service with the nodes in the network. It is also anticipated to resolve the limitations and to significantly improve the performance of other wireless networks. The architecture of WMN [2] is composed of three different network elements and these are (a) Network Gateway (NG) (b) Access Points (AP) or Mesh Routers (MR) and (c) Mobile Nodes (MN). A typical WMN can have a hierarchical structure of three levels of these network elements. At the top level, there are the IGW or the gateway nodes that are directly connected to the wired network. The second level of hierarchy consists of nodes called APs / MRs that forward each other’s traffic in multi-hop fashion towards the IGW. These MRs form the backbone of a WMN and are relatively static. The lowest level of hierarchy is the Mobile Clients / Nodes or the end users connected to the MRs for accessing the wired network services. Usually, most of the traffic in WMNs [3] is oriented towards the Internet, which increases the traffic load on certain paths leading towards the IGW. As the IGWs are responsible for forwarding all the network traffic, they are likely to become potential bottlenecks in WMNs. The high concentration of traffic at a gateway leads to saturation which in turn can result in packet drops due to potential buffer overflows. The packet dropping at the IGWs is not desirable and it makes WMN inefficient because already it had consumed a lot of network resources en route from source to the IGW. Thus, to overcome congestion, the traffic load has to be balanced over different IGWs. The term load balancing refers to optimization of usage of network resources by transferring traffic from congested links to less loaded parts of the network based on knowledge of network state. In a WMN, load balancing is the best approach to increase network throughput and to reduce congestion [4]. Though the load balancing in WMN [5] is critical issue but it is an important concern to utilize the network capacity efficiently. The effects of unbalanced load include gateway loading, center loading, and bottleneck node formation. As the gateway nodes connect the WMN to the external Internet, the traffic aggregation at the gateway nodes creates load imbalance at certain gateways which in turn results in congestion, packet loss. Also the backhaul connection to the external network may become bandwidth constrained. Hence, load balancing across gateways in a WMN is important to improve the bandwidth utilization and network scalability. Though WMNs [3] provide a cost effective way of deploying a network and providing broadband Internet access but it is limited by two main resources: bandwidth and network capacity. The important problems resulting from load imbalance in WMNs are: (a) relay induced capacity degradation, (b) overloading of certain gateway nodes, (c) overloading of WMN nodes closer to the center of the network, and (d) formation and overloading of bottleneck nodes. The remaining of the paper is organized as follows: in section II, a brief description about gateway level load 2014 Fourth International Conference on Communication Systems and Network Technologies 978-1-4799-3070-8/14 $31.00 © 2014 IEEE DOI 10.1109/CSNT.2014.45 193 2014 Fourth International Conference on Communication Systems and Network Technologies 978-1-4799-3070-8/14 $31.00 © 2014 IEEE DOI 10.1109/CSNT.2014.45 193