Access Point Deployment And Topology Control In Wireless Mesh Networks

With the development of wireless communication, wireless network becomes more andmore popular in our lives. Over the last decade, the paradigm of Wireless Mesh Networks(WMNs) has matured to a reasonably commonly understood one. WMNs consist of meshrouters and mesh clients, where mesh routers have minimal mobility and form the backboneof WMNs. They provide network access for both mesh and conventional clients. Theintegration of WMNs with other networks such as the Internet, cellular, IEEE802.11, IEEE802.15, IEEE802.16, sensor networks, etc., can be accomplished through the gateway andbridging functions in the mesh routers. Mesh clients can be either stationary or mobile, andcan form a client mesh network among themselves and with mesh routers. WMNs areanticipated to resolve the limitations and to significantly improve the performance of ad hocnetworks, wireless local area networks (WLANs), wireless personal area networks (WPANs),and wireless metropolitan area networks (WMANs). They are undergoing rapid progress andinspiring numerous deployments. WMNs will deliver wireless services for a large variety ofapplications in personal, local, campus, and metropolitan areas.Despite recent advances in wireless mesh networking, many research challenges remainin all protocol layers. The broadcast nature of wireless signals creates interference, which is amajor issue challenging the performance of WMNs. On one hand, certain high power levelleads to higher data rate, which will reduce the active time of the link and leave more freetime for other links. On the other, high power transmission causes high interference and MAClayer collisions at other unintended receiver. Various attempts have been made to model theeffects of interference using abstract theoretical models as well as measurement-based models.Adopting the knowledge of interference from such models, researchers have designedprotocols for access point placement, channel/radio assignment, power control, linkscheduling, and routing. Despite energy consumption not being an major issue in WMNs,power control and topology control mechanisms mainly deal with assigning transmissionpower levels to nodes such that the traffic demands are satisfied with better overallthroughput.We mainly study some issues existing in the design stage of WMNs, includingfault-tolerant access point placement with QoS constrain, channel assignment consideringpartially overlapping channels in SINR interference model and QoS constraint topologycontrol. The major work can be summarized as follows. 1. Introduce and analyze the characteristic of the wireless mesh networks. Then, discuss themain challenges in the design stage in wireless mesh networks. The challenges includehow to place minimal number of access points to satisfy full coverage and certain specialrequirements such as quality of service and fault tolerance, how to tackle the limitednumber of orghogonal channels by using partially overlapping channels, and how togenerate a valid topology to improve system throughput. For such problems, we givesome effective solutions.2. Give some summary on the main research problem on wireless mesh networks. What'smore, we introduce and analyze some related work. We also explain the interferencemodel and performance and capacity analisis in wireless networks.3. Propose a heuristic method to tackle the issues of fault tolerance access point (AP)placement with QoS constraint in wireless local area networks. The problem of ourconcern is: given a set of users and their traffic demands in a region, our task is to placeminimal number of APs such that one AP fails, the clients affected by the failed AP canstill receive acceptable service from the rest of the APs. We propose a three-phase methodto solve this problem. In the first phase, we aim to place minimal number of APs in whicheach client can connect to one AP and its degrading traffic demand can be met. In thesecond phase, we continue to add APs to meet per-client normal traffic demand. In thelast phase, we palce more APs to tolerate the failure of APs palced in the first phase.Simulation results show the efficiency of our proposed method.4. Base on the signal to interference noise ratio model, by using partially overlappingchannels, proposed a greedy method to tackle the issues of channel assignment of basicservice sets in wireless local area networks in a detail way. Due to the limited number oforthogonal channels in IEEE802.11b/g standards, using only non-overlapping channelscannot completely eliminate the interference. Recent studies indicate that we can improvethe full-range channel utilization and the network throughput by properly utilizing thepartially overlapping channels. However, there are only limited studies of channelassignment algorithms for partially overlapping channels. In this paper, we investigate theproblem of partially overlapping channel assignment to improve the performance of802.11wireless networks based on SINR model. Using the SINR model, we deduce adirect relationship between maximizing system throughput and minimizing totalinterference when partially overlapping channels are employed. After that, we propose agreedy method to minimize the total interference for throughput maximization. Weevaluate our algorithm through extensive simulations and compare its performances with those of the state-of-the-art.5. Study the scenario of a wireless mesh backbone network, considering rate adapatation,how to generate network topology and improve systeme throughput using power controland topology. We propose a collision load balancing topology control stategy. Given a setof mesh routers in a plane, some of which have wired connection to the Internet and arecalled gateway nodes. Each node carries end users' traffic demand and there is thresholdof delivery ratio α0for all nodes in the system. Our task is to construct a routing forest (aset of trees rooted from the gateway nodes), configure the transmission power of eachnode, such that the constructed network topology can meet the delivery threshold for allnodes and the system throughput is maximized. We consider transmission rate adaptationin this paper, i.e., the data rate over a link depends on the signal strength at recipient node.Combining with power control and data adaptation, we propose a collision-loadbalancing method to construct the routing trees. Our method has two steps. In the firststep, all nodes's delivery ratio is set to α0and we construct the routing trees such that thesystem interference is minimized. In the second step, we distribute the remaining systemcapacity, in terms of delivery ratio, evenly to the nodes close to the gateways, becausethese nodes' traffic needs less relay times. Our experiments show that power control andrate adaptation scheme can improve the system performance more effectively. Simulationresults have also shown that our method can significantly improve the networkthroughput, compared with shortest path tree (SPT) method.6. Conclude this work and disscuss the future work.