Abstract- The Wireless Mesh Network backbone is usually comprised of stationary nodes that are not uniformly distributed and connectivity is traditionally assured by requiring that all nodes use their maximum transceiver powers. The use of maximum transceiver powers may actually not be necessary to ensure network connectivity and this realization has led to the development of two distinct types of network connectivity strategies in the literature. This paper focuses on the network-size-dependent Critical Number of Neighbors (CNN) strategy. An evaluation of several well-known examples of this strategy was undertaken on an indoor test-bed. The evaluation indicates that non- uniform node placement necessitates the use of greater CNN values and that the scheme devised by Xue and Kumar was most suitable when both network connectivity and transceiver power consumption were considered. The implications of the use of network-size- dependent connectivity strategies in the real world are also considered. Index Terms—Wireless Mesh Networks, connectivity, CNN, power control, topology control I. INTRODUCTION Infrastructure Wireless Mesh Networks (WMNs) are a sub-class of ad hoc networks that possess a two-tier architecture consisting of an access and a backbone network. Client devices connect to the mesh backbone which is typically self-organizing and self-configuring and backbone nodes are usually co-located with existing infrastructure in real-world deployments [1], [2]. These backbone nodes, comprising Mesh Points (MPs), Mesh Access Points (MAPs) and Mesh Portals (MPPs), collaborate amongst themselves to maintain network connectivity and deliver traffic to the intended destinations (see Fig. 1). Despite the stationary nature of the infrastructure WMN backbone, maintaining network connectivity is made difficult by the transient nature of wireless links, making them unreliable when deployed [3], [4], [5]. Traditionally, network connectivity is assured by ensuring that each device in the WMN backbone utilizes its maximum transceiver power. The disadvantages of this approach are the high levels of interference, increased contention for the transmission medium, a reduction in network capacity and unnecessary transceiver power consumption. Network connectivity determines the network topology Fig. 1. - Infrastructure WMN Architecture [12] which has been shown to have significant influence on the ability of a network to provide QoS support. As a result several studies have been undertaken to devise strategies for optimal network connectivity and the results of these studies have the potential to be utilized in Topology Control [6] schemes. These connectivity studies have demonstrated that the various strategies employed possess the ability to create interference-efficient network topologies [7], [8] as well as providing route redundancy in some cases [9], [10], [11]. To the best of our knowledge a test-bed evaluation of the ability of previously proposed connectivity strategies’ ability to maintain network connectivity when nodes are not uniformly distributed is yet to occur. The use of network connectivity strategies in real-world WMN backbones is dependent upon the backbone nodes being capable of transceiver power control. This capability is required so that transceiver powers can be adjusted according to the connectivity strategy being employed. In this paper various connectivity strategies based on the variable Critical Neighbor Number (CNN) approach [13], [11], [14], [6] are evaluated on an indoor test-bed consisting of Linksys WRT54GL devices. These devices are popular WMN backbone nodes and the test-bed evaluation indicates that most of the connectivity strategies being evaluated were able to achieve transceiver power savings but that some strategies achieved these savings at the expense of network connectivity. Those strategies that did achieve network connectivity were subsequently tested to determine the time taken to converge upon a connected network topology. These convergence results indicate that there exists an inverse relationship between the power savings achieved and the network convergence time. The connectivity A Test-Bed Evaluation of Connectivity Strategies for Infrastructure Wireless Mesh Networks Pragasen Mudali 1 , Thulani C. Nyandeni 2 , Ntsibane Ntlatlapa 3 and Matthew O. Adigun 1 Department of Computer Science University of Zululand 1 , Private Bag X1001, KwaDlangezwa 3886 Tel: +27 35 9026012 and Telkom SA 2 and Meraka Institute CSIR 3 email: pmudali@pan.uzulu.ac.za 1