Yan Zhang / Wireless Mesh Networking AU_C Final Proof page i pm .. Wireless Mesh Networking: Architectures, Protocols and Standards networks—a security whitepaper. aracer.mobi Tropos_. Wireless Mesh Networking: Architectures, Protocols and Standards. (Wireless aracer.mobi?tp=&ar SANTASHIL. Issues in Architecture and Protocol Design for Wireless Mesh Networks Chapter 1, IEEE a standard for broadband wireless access in metropolitan area .
|Language:||English, Spanish, Indonesian|
|Genre:||Academic & Education|
|Distribution:||Free* [*Registration needed]|
Abstract— WMN, the Wireless Mesh Network is a promising technology in providing the last mile overview (i.e., fundamental concept, technology, standards, etc.) . which is Hybrid Wireless Mesh Protocol (HWMP) for its routing protocol. Wireless mesh network - Wikipedia, the free encyclopedia Wireless mesh architecture is a first step towards providing cost effective and . Standard autoconfiguration protocols, such as DHCP or IPv6 (aracer.mobi aracer.mobi), in IEEE Wireless Communications, vol 10, 5 pp Wireless Mesh Networks Architectures and Protocols, Springer, ▫ A. Ting, and D. . Standard technologies: IEEE , , 5. Specific PHY.
A common set of commissioning methods is provided, including Touchlink, a method of proximity commissioning. Zigbee 3. Any Zigbee router node can subsequently provide the network key to joining nodes.
Nodes adopt whichever security method is used by the network they join. Zigbee also handles the dynamic behavior of these networks with nodes appearing, disappearing and re-appearing in the network and allows orphaned nodes, which result from the loss of a parent, to re-join the network via a different parent. The self-healing nature of Zigbee Mesh networks also allows nodes to drop out of the network without any disruption to internal routing.
The backward compatibility of Zigbee 3. The Smart Energy profile is also compatible with Zigbee 3. OTA upgrade is an optional functionality that manufacturers are encouraged to support in their Zigbee products.
Mesh Networks A key component of the Zigbee protocol is the ability to support mesh networking. Skalli, deals with the problem of assigning channels to radio interfaces in a multi-channel and multi-radio wireless mesh backbone network.
The key challenges associated with the channel assignment problem are outlined and a survey on the existing channel assignment schemes is provided. The objective of a channel assignment strategy is to ensure efficient utilization of the available channels e. However, since these two requirements are conflicting with each other, the goal is to achieve a balance between these two. The major constraints which need to be satisfied by a channel assignment scheme include: Also, a channel assignment scheme should take the amount of traffic load supported by each mesh node into consideration.
Optimal channel assignment in an arbitrary wireless mesh backbone is an NPhard problem similar to the graph coloring problem. The existing channel assignment schemes in the literature are, therefore, mostly heuristic based. These schemes can be classified into three categories: Fixed assignment schemes assign channels to the radios either permanently or for a long time interval.
With dynamic channel assignment, the radios can frequently switch from one channel to another. Hybrid channel assignment strategies apply a fixed assignment for some radios and a dynamic assignment for other radios. Fixed channel assignment schemes can be further classified into two categories: In CCA, all the radios in all of the mesh nodes are assigned the same set of channels.
In VCA, radios of different nodes are assigned different sets of channels.
The authors have described a number of such VCA schemes. With dynamic channel assignment, when two mesh nodes need to communicate with each other, they need to switch to the same channel. The key challenge in this case is how to coordinate the switching decisions. The authors have described a number of dynamic channel assignment schemes. Hybrid assignment strategies are attractive since they allow for simple coordination algorithms as for the fixed assignment schemes and also provides the flexibility XIV Preface of dynamic channel assignment.
The authors have described two such hybrid channel assignment schemes. The key issues considered in the design of the existing channel assignment schemes are network connectivity, constraint on topology, interference minimization, effects of link revisits, traffic awareness, switching overhead for dynamic and hybrid schemes , and control philosophy i.
Considering these factors, the authors provide a qualitative comparison among the different schemes. Xue, Y. Cui, and K. Nahrstedt, presents a generalized theoretical framework for resource allocation and transmission rate control in wireless mesh networks.
The objective of this framework is to achieve optimal resource utilization and rate fairness among flows on an end-to-end basis. Based on this theoretical framework, the authors also present a price-based distributed algorithm for resource allocation which converges to the globally optimal solution.
The resource allocation problem is first formulated as an optimization problem for an abstract network model consisting of a set of resource elements e. The objective is to maximize the aggregated utility i. Different fairness models such as weighted proportional fairness and max-min fairness can be implemented through the appropriate choice of the utility function.
The solution of the optimization achieves both optimal resource utilization i. Based on the Lagrangian form of the optimization formulation, a price-based decentralized solution can be obtained which depends on local decision of each resource element and exchange of control signals among them.
The authors show that for a multihop wireless mesh backbone network, a resource element is a facet of the polytope defined by the independent sets of the conflict graph of this network. It can be approximated by a maximal clique of the contention graph which basically represents a maximal distinct contention region in the network.
The resource constraints in the network can then be represented by the achievable channel capacities in all of the maximal cliques in the contention graph. Subsequently, the end-to-end rate allocations can be obtained for the flows. For distributed implementation, a flow adapts its rate as a function of price it pays to all resource elements, where the price for a resource element is a non-negative, continuous, and increasing function of the total traffic served by that resource element. The authors show that the rate adaptation algorithm is stable and at the equilibrium each flow maximizes its utility.
Sayegh, T. Todd, and M. Resource allocation in such a node involves assigning solar panel or wind turbine size, and battery capacity, and this resource allocation depends on the geographic location of the node.
The charge controller disconnects the battery from the power source to protect it from under- and over- charging. Specifically, when the residual battery energy falls below the maximum allowed level of discharge, the charge controller disconnects the node load and the node then experiences a radio outage. In a hybrid configuration, both solar panel and wind turbine are used. The authors investigate the short-term statistics of the energy available from solar panel and wind turbines at two different locations, namely, Toronto, Ontario and Phoenix, Arizona.
However, the short-term statistics may not be sufficient to assess the optimal dimensioning of the power source in the mesh node. The long-term statistics would be required instead. Examples of long-term statistics show that performance metrics such as radio outage probability for the wind source and the solar source depends on the seasonal correlation between solar power and wind power in a geographic location. The desired level of sustainability of a given hybrid system for the different geographical locations can be obtained by properly choosing the wind turbine and battery sizes.
To minimize the total cost of a hybrid node i. This optimization model is solved numerically. To this end, the authors show that power saving at mesh access points can greatly reduce the cost which is almost proportional to the power consumption in the node. Badia, A. Erta, L.
Lenzini, and M. Zorzi, presents a comprehensive survey on the state-of-the art of routing and link scheduling in wireless mesh networks. As has been mentioned before, for a wireless mesh network, the objective of a routing algorithm is to discover efficient paths to obtain high system throughput. Link scheduling at the medium access control layer is used to activate XVI Preface the communication links with an objective to ensuring the desired level of network connectivity under interference constraints.
The interference models, which are particularly important when designing link scheduling or activation and routing algorithms, can be of three types - physical, protocol, and measurement-based interference models. With a physical interference model, the feasibility of simultaneous link activations is determined by the SINR at the receivers. Note that, the packet error rate at a receiver is a monotonically decreasing function of SINR. With a protocol interference model, simultaneous transmissions result in incorrect decoding of a received packet.
The measurement-based interference model takes an a priori approach to interference characterization. Designing a framework for joint scheduling and routing which considers the network requirements, resource constraints e. The authors propose a graph-based approach to design a framework for joint link scheduling and routing through link activation. In this framework, the radio transceiver constraints e. The interference is characterized by a physical interference model which is more accurate than that under protocol interference models from the viewpoint of theoretical analysis of wireless mesh networks.
The authors assume a centralized space time division multiple access STDMA scheme to obtain an efficient transmission scheme through link activation. The mesh access point nodes in the mesh backbone network finds the link activation patterns in a centralized manner and communicates it with the other nodes. The authors also carry out some numerical investigations on the performance of the proposed framework for different interference models. Koksal, presents a comparative study among seven different link cost metrics for routing in wireless mesh networks.
The cost metric for a link refers to the cost of forwarding a packet along that link. The considered link cost metrics are: The traditional hop count-based routing i. Per-hop round trip time is a delay-based link cost metric, which is calculated by a mesh node as the exponentially weighted moving Preface XVII average of the RTT samples for each of its neighbors.
This metric takes into account the factors such as queueing delay, channel quality, and channel contention. However, since RTT varies with varying load, using this routing metric may lead to route instability due to the self interference effect. With this routing metric, the optimal path assignments may change more frequently compared to the hop count, which may result in reduced network throughput.
Also, this metric responds to channel variability at time scales longer than tens of packets. The PktPair metric is obtained as the difference between the times of reception of two successive packets. Therefore, it does not take into account the queueing and processing delay at a node.
Although it suppresses the route instability effect to some extent, the overhead associated with it is higher than that due to per hop RTT. The quantized loss rate is based on the end-to-end packet loss probability.
This metric does not take the link bandwidth into account, and therefore, low bandwidth paths could be chosen for routing. ETX for a wireless link refers to the estimated expected number of transmissions required to transfer a packet successfully over that link. This metric depends only on the link level packet errors due to channel impairments, and therefore, the effects of self interference is reduced. ETX can improve the throughput performance significantly compared to the hop count metric, however, it may perform poorly under highly variable and bursty error situations.
This metric is a function of the mean and the variance of the bit error probability summed over a packet duration. It offers a higher throughput performance compared to the ETS metric. However, the main drawback of this metric is the complexity of estimation of the mean and variance of bit error probability.
Also, estimation error may impact its performance significantly. It is used to find routes which satisfy certain desired end-to-end performance e. The authors also present a unified geometric framework to compare the different routing metrics.
This framework combines the mean and standard deviation of the bit error rate process. In this framework, it is possible to define a feasible region using which links can be selected to achieve the desired routing performance.
Baiamonte, C. Casetti, C. Chiasserini, and M. Specifically, the authors consider the problem of designing efficient relaying schemes based on the cross-layer design principles which take into account the quality of the wireless links in an XVIII Preface For traffic flow from a mesh gateway to wireless mesh nodes, the authors present two schemes for packet forwarding, namely, the split queues SQ approach and the access category AC approach.
With the former approach, two queues are maintained at each node for relay traffic and local traffic. With the latter approach, several queues are implemented at the MAC layer, each of which is associated with a priority level e. Prioritizing relay traffic over local traffic provides an incentive to the nodes to act as relays.
Simulation results for a network topology with single and multiple relays serving TCP and UDP flows show that the AC approach can provide significant gain in throughput while the SQ approach can provide very high fairness in throughput. The authors present a fair relay selection algorithm FRSA which is an extension of the optimized link state routing OLSR protocol designed for wireless ad hoc networks.
OLSR is a table-driven and a proactive protocol which exchanges topology information periodically with other nodes in the network. The route from a given node to any destination node in the network is formed by relay nodes. A relay node announces to the network that it has reachability to the nodes which have selected it as the relay node. In FRSA, each node performs a relay quality-aware routing to its two-hop neighborhood.
Gkelias and K. Leung, discusses the research challenges associated with the deployment of multiple antenna technologies in wireless mesh networks.
In particular, the authors focus on the design of medium access control and routing algorithms in wireless mesh networks employing smart antenna technology. Multiple antenna technology includes fixed beam antenna techniques, adaptive antenna techniques, and multiple-input multiple-output MIMO coding techniques which can be highly beneficial to improving overall performance of wireless mesh networks. However, employment of multiple antenna or smart antenna techniques in a wireless mesh networking environment gives rise to unique problems such as deafness, hidden and exposed terminals, and multi-stream interference.
Novel medium access control and routing protocols need to be designed to address the above problems. The authors first describe the wireless mesh network and channel characteristics considering different propagation scenarios, interference characteristics in different scenarios, and other constraints such as the limitations in total effective radiation power. Then an overview of the different smart antenna techniques is provided. Two basic types of smart antennas, namely, directional antennas fixed beams and adaptive antenna arrays, are considered.
Directional antenna techniques, which include switched-beam antennas, steered-beam antennas or dynamically phased array antennas , can provide high SINR gain in presence of strong line-of-sight component, Preface XIX however, their performances degrade in multi-path environments. Adaptive antenna techniques, which include adaptive antenna arrays and MIMO techniques, can provide high gain in the direction of desired signals and nulls in the direction of undesired signals i.
In particular, the MIMO techniques can exploit the multi-path fading effects to enhance the transmission rate i. One of the major issues related to the use of multiple antenna or smart antenna techniques in wireless mesh networks is to mitigate the deafness problem. This problem arises due to the use of directional antennas when a transmitter fails to communicate with its intended receiver. However, deafness can be also exploited in some cases to mitigate interference.
Again, in presence of directional antennas, unsuccessful transmissions due to packet collision and deafness need to be treated differently at the higher layers. In a MIMO-based wireless mesh network, the medium access control protocol should use the optimal number of simultaneous transmissions, allocate appropriate number of streams per transmitterreceiver pair, and perform power allocation accordingly.
Also, the tradeoff between multiplexing and diversity gain should be taken into account. If the higher layer protocols are not carefully designed, the multiple antenna techniques can have negative impact on the overall network performance.
The authors then discuss several distributed medium access control protocols for multiple antenna-based multihop wireless networks. The interactions between medium access and routing protocols in presence of smart antennas have been evaluated in some works in the literature. These works primarily focused in improving network connectivity. Design and implementation of efficient quality of service QoS -aware routing protocols which exploit the multiple antenna techniques is a grand research challenge.
Zhang, Z. Wang, S. Das, and M. Hassan, addresses the security issues in wireless mesh networks. The main challenges for securing wireless mesh networks arise due to the requirements of authentication, secure routing, secure location information of mesh routers , and to defend against virus attacks. Authentication is required to distinguish malicious information from legitimate information. An authentication mechanism is generally implemented with the help of public key infrastructure PKI and certification authority CA.
With the PKI mechanism, each user has a pair of cryptographic keys: A message encrypted with the public key which is known to all the users can only be decrypted by using the corresponding private key, and vice versa.
It is assumed that the signed certificates by the CA are globally trusted in the network. Due to the absence of any pre-established trusted network infrastructure in wireless mesh networks, distributed CA schemes are desirable. The authors describe a number of such CA schemes. The routing protocols for a wireless mesh network are vulnerable to both external and internal attacks.
External attackers can inject fabricated routing information into the network or maliciously alter the contents of routing messages. An internal attack is launched from within a node when an attacker gains full control of the node. To prevent external attackers from sending fabricated routing information, cryptography-based authentication methods incorporated in the routing protocols can be used. The authors describe several of such schemes. Also, several possible approaches to detect and counter measure the internal attacks to routing protocols are discussed.
Securing the location information of wireless mesh routers is crucial for certain type of routing schemes e. Two methods for securing location information are generally used - correctly computing the location information and verifying the location claims.
The authors review several works based on these two methods. Computer viruses also pose threats to security in wireless mesh networks. There have been research efforts towards modeling the virus propagation problem in wireless networks.
Epidemic theory used in Biology is one popular technique used to investigate the virus spreading problem. Two schemes which use Epidemic theory to model the propagation of viruses and compromised nodes, respectively, are discussed. The authors also outline a number of security-related research issues in wireless mesh networks. These include securing the medium access control protocols, defending against denial of service DoS attacks at the different layers in the protocol stack, designing cross-layer framework for self-adapted security mechanisms, customizing the security schemes based on the type of network in a heterogeneous wireless mesh environment , and trust establishment and management.
All of these issues represent fertile areas of future research in wireless mesh networks. Conclusion We have provided a summary of the contributed articles in this book.
We hope this summary would be helpful to follow the rest of the book easily. We believe that the readers will find the rich set of references in each of the articles very valuable.
We would like to express our sincere appreciation to all of the authors for their excellent contibutions and their patience during the publication process of the book. We hope this book will be useful to both researchers and practitioners in this emerging area. Sandhu O.
Gungor1 , E. Natalizio2 , P. Pace3 , and S. It is gaining significant attention as a possible way for Internet service providers ISPs and other end-users to establish robust and reliable wireless broadband service access at a reasonable cost. WMNs consist of mesh routers and mesh clients as shown in Fig. In this architecture, while static mesh routers form the wireless backbone, mesh clients access the network through mesh routers as well as directly meshing with each other.
Different from traditional wireless networks, WMN is dynamically self-organized and self-configured. In other words, the nodes in the mesh network automatically establish and maintain network connectivity.
This feature brings many advantages for the end-users, such as low up-front cost, easy network maintenance, robustness, and reliable service coverage. In addition, with the use of advanced radio technologies, e. Moreover, the gateway and bridge functionalities in mesh routers enable the integration of wireless mesh networks with various existing wireless networks, such as wireless sensor networks, wireless-fidelity Wi-Fi , and WiMAX .
Consequently, through an integrated wireless mesh network, the end-users can take the advantage of multiple wireless networks. Some of the benefits and characteristics of wireless mesh networks are highlighted as follows: In WMNs, the wireless mesh routers provide redundant paths between the sender and the receiver of the wireless connection.
This eliminates single point failures and potential bottleneck links, resulting in significantly 2 V. Gungor et al. Network robustness against potential problems, e. Therefore, by utilizing WMN technology, the network can operate reliably over an extended period of time, even in the presence of a network element failure or network congestion. Recently, the main effort to provide wireless connection to the end-users is through the deployment of To assure almost full coverage in a metro scale area, it is required to deploy a large number of access points because of the limited transmission range of the APs.
The drawback of this solution is highly expensive infrastructure costs, since an expensive cabled connection to the wired Internet backbone is necessary for each AP. On the other hand, constructing a wireless mesh network decreases the infrastructure costs, since the mesh network requires only a few points of connection to the wired network.
Currently, the data rates of wireless local area networks WLANs have been increased, e. Although the data rates of WLANs are increasing, for a specific transmission power, the coverage and connectivity of WLANs decreases as the end-user becomes further from the access point.
On the other hand, multi-hop and multi-channel communications among mesh routers and long transmission range of WiMAX towers deployed in WMNs can enable long distance communication without any significant performance degradation.
Wireless mesh networks are dynamically self-organized and self-configured. In other words, the mesh clients and routers automatically establish and maintain network connectivity, which enables seamless multi-hop interconnection service. For example, when new nodes are added into the network, these nodes utilize their meshing functionalities to automatically discover all possible routers and determine the optimal paths to the wired Internet . Furthermore, the existing mesh routers reorganize the network considering the newly available routes and hence, the network can be easily expanded.
In this chapter, we present a survey of recent developments in the protocols and architectures for WMNs and discuss the opportunities and challenges of WMNs. The motivation of this chapter is to provide a better understanding of wireless mesh network technology that can ensure heterogeneous application requirements.
Consequently, our aim is to present a structured framework for the end-users who plan to utilize WMNs for their applications and hence, to make the decision-making process more effective and direct. The rest of the chapter is organized as follows. In Section 1.
Physical testbeds and standardization activities in WMNs are explored in Section 1. Finally, the conclusions are stated. An illustration of wireless mesh network architecture. Mesh routers are resourcerich nodes equipped with high processing and memory capabilities, while mesh clients have limited memory and computational power.
Unlike a traditional ad hoc network, which is an isolated selfconfigured wireless network, the mesh network architecture introduces a hierarchy with the implementation of dedicated and power enabled mesh routers. In this integrated network architecture, some of the mesh routers are also called as gateways, which are special wireless routers with a high-bandwidth wired connection to the 4 V. More specifically, mesh routers contain advanced routing functionalities to support mesh networking.
This feature of mesh routers is realistic, since mesh routers are fixed nodes, with no constraints on power supply since they are assumed to be connected to power lines , with multiple wireless interfaces built on either the same or different wireless access technologies. Different from mesh routers, mesh clients can be mobile nodes, which typically run on batteries. Thus, power usage of mesh clients should be limited. This can be achieved by means of reduced radio functions, e.
The target technology for both mesh routers and mesh clients is the IEEE The main reason is the widespread availability of However, to leverage on this opportunity, the modifications required by mesh routers and mesh clients should be aware of the existing hardware constraints and limitations.
In WMNs, a careful planning of hardware resources needs to be devised in terms of the position and the number of wireless interfaces, and the technology limitations.
Also, how to best place mesh routers impacts the network capacity and topology, and thus, needs to be investigated. A sophisticated network management tool needs to be developed for both mesh routers and mesh clients to dynamically establish connections between them and to closely follow the dynamics of traffic load and users mobility. It is necessary to design a low-cost method to integrate the IEEE However, which mesh router shall have additional interfaces is also part of the network planning issue.
Although there exist recent advances in the mesh networking technology, many research problems still need to be resolved: The critical factors influencing the performance of WMNs can be summarized as follows: Recently, many solutions have been proposed to improve the capacity of WMNs.
However, the complexity and the cost of these technologies are still too high to be widely accepted for the commercialization. Therefore, all these advanced wireless radio technologies require a revolutionary design in the communication protocol suite in order to facilitate the deployment of WMNs and the commercialization of the products.
Existing networking technologies have limited capabilities of integrating different wireless networks. Denial of service attacks and intrusions in WMNs can cause severe damage to the operation of the deployed network.
Although there exist many security schemes proposed for wireless local area networks and ad hoc networks, most of these security solutions are either not practical or showing poor performance in WMNs because of the lack of a centralized trusted authority to distribute a public key in the WMN architecture. Consequently, there is a need for new security schemes ranging from efficient encryption and authentication mechanisms to secure key distributions, and intrusion detection mechanisms.
The deployed mesh network must be able to deal with large network topologies without increasing the number of network operations exponentially. In addition, the network performance should not degrade as the number of hops between the sender and the receiver increases. To provide the scalability in WMNs, there is a need for scalable MAC, routing and transport layer protocols with minimum overhead. The network services that are provided by WMNs vary from reliable file transfer to real-time multimedia, such as live video streaming.
Thus, in addition to traditional network throughput and communication latency metrics, more comprehensive performance metrics, such as delay jitter, aggregate and per-node fairness, and packet loss ratios, need to be considered by the developed mechanisms. In WMNs, to eliminate the single point failures and potential bottleneck links, the wireless back- 6 V. However, the topology and connectivity of the network can vary frequently because of the route failures and energy depletions6.
Therefore, to take all the advantages of autonomous mesh connectivity, efficient network self-configuration, topology control and power management algorithms are required. To support mobile mesh clients in WMNs, it is necessary to design advanced physical layer and networking techniques, which adapt to the fast fading conditions commonly associated with the mobile users. In addition to these advanced techniques, low latency handover and location management algorithms are also required to improve the quality of service during mobility.
To monitor the overall network performance and maintain the network operation, flexible and scalable network management capabilities are required for WMNs.
The primary network management capabilities of the WMNs include: Recently, several commercially interesting applications for broadband wireless services have been deployed based on the wireless mesh network architecture.
However, since numerous applications can be supported by the WMNs, it is infeasible to have a complete list of them. Recently, several Internet Service Providers ISPs deploy wireless mesh networks WMNs to enable broadband wireless services in urban, suburban, and rural environments  and .
These WMN deployments bring significant advantages over traditional wireless networks, including extended network coverage, high speed, and cost-effective network installation. Therefore, the deployments of WMNs are also expected to grow with the increase in demand for broadband wireless Internet access.
Wireless mesh networks appear to be one of the most promising solutions to address the needs of law enforcement agencies and city governments, such as the police, fire departments, first responders, and emergency services. Currently, several mesh networks are operating to provide mobility support, reliability, flexibility, and high bandwidth for public safety applications , ,  and . However, the recent field trials and experiments with existing communication technologies show that the performance of WMNs is still below what they are expected to be.
Consequently, there is a need for the development of large-scale physical test-beds and novel communication protocol suites for WMNs. In a building, the operation of various electrical devices, including ventilation and air conditioning HVAC systems, power, light, elevator, etc. Traditionally, all these operations are realized using wired networks, which is very expensive due to the installation and maintenance costs.
In this context, wireless mesh networks can offer efficient and cost-effective solutions for advanced building automation systems. Equipment failures, lightning strikes, accidents, and natural catastrophes all cause power disturbances and outages and often result in long service interruptions . WMNs can provide an economically feasible solution for the wide deployment of high speed wireless communications for electric utility automation applications, such as real-time grid and equipment monitoring, incipient fault detection and identification, and wireless automatic meter reading.
Currently, there exist several peer-topeer P2P networking protocols for information sharing on the Internet. Therefore, in order to support P2P applications, new well designed protocols need to be integrated into the application layer.
Recently, various public transportation companies and the government agencies are interested in practical networking solutions to realize the information delivery system controlling several transportation services . In this regard, wireless mesh networks WMNs can provide flexible wireless networking solutions to intelligent transportation systems. With the use of WMNs, the problems of transportation congestion can be addressed and transportation security and safety can be improved.
To provide strict quality of service QoS requirements of the applications and to create application protocols for managing distributed information sharing in WMNs, the protocols in the lower layers need to work interactively with the application layer. This requires a cross-layer approach through information sharing among application, transport, routing, medium access control MAC and physical layers. In this way, the deployed WMN can be self-adaptive to network dynamics and meet end-to-end real-time deadlines of the applications.
To enable large-scale WMNs and to realize fully integrated and cooperative wireless networking solution, new and commercially interesting applications need to be studied based on the exclusive features and advantages of the WMNs. In this way, this new technology can be made very attractive for both consumers and service providers. Novel application protocols that incorporate the use of pricing as an incentive mechanism to encourage private and self-interested nodes to participate in a public wireless mesh network need to be studied.
For example, the Internet access can be considered as a service, and hence access points are the service sellers. In this respect, any downstream wireless mesh nodes may download this service, for its own consumption, or for reselling it to other downstream nodes obtaining a fair revenue.
In this section, we explain existing transport layer protocols with a focus on ad hoc networks, since WMNs share common features with ad hoc networks in spite of their differences. In any case, it is useful to keep in mind that efficient transport protocols are needed for nonreal-time and real-time traffic for satisfying different QoS requirements in WMNs. TCP-Based Solutions Most of the wireless transport protocols proposed in the literature or in use today are enhancements of TCP, which is originally designed to work in the wired Internet .
The shortcomings of TCP in wireless ad hoc networks have been investigated in , , , , , . In this section, we briefly discuss some of the significant drawbacks of TCP-based solutions in the context of wireless mesh networks. More specifically, we categorize the discussion based on the following characteristics of TCP-based solutions: In WMNs, The route failures and consequent route changes affect the congestion control performance of TCPbased solutions significantly.
Whenever a route changes, a TCP-based solution employs a slow start mechanism to probe for the available throughput capacity. This mechanism does not allow to increase the rate aggressively, since every connection takes several round-trip-time RTT periods before reaching its effective bandwidth value, spending a considerable portion of its lifetime in the probe state.
This behavior leads to an under-utilization of network resources, especially for dynamic wireless networks. Based on the wellknown square root formula , when TCP losses occur primarily due to link errors, the TCP throughput of each connection can be represented as a function of p and RTT: This formula also shows another inefficiency of TCP-based solutions, i.
In WMNs, end-to-end congestion detection and control can be imprecise because of the inaccurate estimations of the RTT and the dynamic nature of the wireless channel. In multi-hop wireless mesh networks, link failures are frequent and happen either due to the nodes moving out of range of each other, or due to heavy contention, which is perceived as a link breakage on repeated failures to deliver a packet.
These breakages lead to route failures, which then result in frequent route re-computations. Based on the TCP drawbacks, which are revealed through many performance evaluation studies, several transport layer solutions have been proposed in the literature for wireless ad hoc networks.
All these solutions propose to solve the problems by improving TCP with additional functionalities, modifications, or getting support from lower layers. In this section, we list various enhanced TCP protocols by addressing the proposed solutions to the classical TCP problems on wireless networks. In , link level protection and ACKing mechanism were advocated to improve the TCP performance over wireless ad hoc networks. In , the problems of TCP in dynamic multihop wireless networks were determined and additional mechanisms at media access and routing layers were proposed to improve TCP performance.
The 10 V. In , a transport layer solution ATCP was proposed, which introduces a thin layer between the transport and underlying routing layers to improve TCP performance by putting TCP into persist mode whenever the network gets disconnected or there are packet losses due to high bit error rate. In , an adaptive pacing mechanism TCP-AP was developed for wireless multi-hop networks in order to avoid bursty packet transmissions. It is important to note that all these protocols are based on end-to-end rate adjustment and congestion control mechanisms and require a fine-grained end-to-end communication between the source and the destination.
Therefore, they may experience significant network inefficiency in WMNs due to the dynamic characteristics of multi-hop wireless environments, end-to-end delay and even obsolete receiver rate feedbacks. Novel Transport Protocols Researchers have developed entirely novel transport protocols for both ad hoc and mesh networks to address the fundamental problems existing in TCP, as argued in the previous sections.
In , the ad hoc transport protocol ATP was proposed for mobile ad hoc networks. The ATP utilizies a rate-based transmission mechanism for rate estimation, and a quick start algorithm for the initial bandwidth estimation. Also, it decouples the congestion related and non-congestion related losses. In this way, the ATP achieves higher performance compared the TCP variants in terms of communication delay, network throughput, and fairness.
Recently, an adaptive and responsive transport protocol AR-TP for WMNs has been proposed in  in order to fairly allocate the network resources among multiple flows, while minimizing the performance overhead.
AR-TP includes both efficient hop-by-hop rate adjustment and reliability mechanisms to achieve high performance reliable data transport in WMNs. Compared to end-to-end rate control schemes, hop-by-hop rate adaptation strategy of the AR-TP protocol enables each router to keep track of dynamic wireless channel conditions in a responsive manner. In addition, with the use of hop-by-hop strategy, the AR-TP can adapt its data transmission rate opportunistically in case of multi-channel WMNs.
Performance evaluation via extensive simulation experiments show that the AR-TP protocol achieves high performance in terms of network throughput and fairness. Transport Protocols for Real-Time Communication In WMNs, a real-time rate control protocol is necessary to meet the end-to-end deadlines of the applications .
This protocol proposes a multimetric joint detection mechanism for TCP-friendly rate control algorithms. However, 1 Architectures and Protocols for Wireless Mesh Networks 11 the performance of the detection mechanism is not satisfactory to deliver real-time multimedia traffic.
RCM does not distinguish between congestion loss and link loss. Once a loss is detected, RCM reduces the sending rate. This behavior makes the protocol power efficient since reducing the sending rate in burst error state may reduce the link corruption.
When there is no loss, a new rate increase mechanism, which takes burstiness of loss into account heuristically, was used. Specifically, if a heavier burst loss is detected, a more aggressive increase is applied in order to recover quickly after large rate reduction. However, the existing TCP-friendly rate control protocols ,  cannot be used in WMNs to support real-time delivery for multimedia traffic since all non-congestion packet losses caused by different reasons are handled in the same way.
Fig 3. Client WMNs 3. Mesh clients can access the network through mesh routers as well as directly meshing with other mesh clients. While the infrastructure provides connectivity to other networks such as the Internet, Wi-Fi, WiMAX, cellular, and sensor networks; the routing capabilities of clients provide improved connectivity and coverage inside the WMN.
The hybrid architecture will be the most applicable case in our opinion. Fig 4. Deployment scenarios that are particularly well suited for WMNs include the following: Due to the recent research advances in WMNs, they have been used in numerous applications.
The mesh topology of the WMNs provides many alternative paths for any pair of source and destination nodes, resulting in quick reconfiguration of the path when there is a path failure. WMNs provide the most economical data transfer coupled with freedom of mobility. Mesh routers can be added incrementally to improve the coverage area.
The obvious problem here is the location of the access point in the home, which may lead to dead zones without service coverage.
These problems can be solved by replacing all the access points by the mesh routers and establishing mesh connectivity between them. This provides broadband connectivity between the home networking devices and only a single connection to the Internet is needed through the gateway router.
By changing the location and number of mesh routers, the dead zones can be eliminated. Fig 5 shows a typical home network using mesh routers. Fig 5. All the traffic in community networking goes through the Internet, which leads to inefficient utilization of the network resources. The last mile of wireless connectivity might not provide coverage outside the home. Community networking by WMNs solves all these problems and provides a cost effective way to share Internet access and other network resources among different homes.
Fig 6 shows wireless mesh networking by placing the mesh routers on the rooftop of houses. There are many advantages to enabling such mesh connectivity to form a community mesh network.
For example, when enough neighbors cooperate and forward each others packets, they do not need individual Internet connectivity; instead they can get faster, cost-effective Internet access via gateways distributed in their neighborhood. Another advantage is that neighbors can cooperatively deploy backup technology and never have to worry about losing information due to a catastrophic disk failure.
Another advantage is that this technology alleviates the need for routing traffic belonging to community networking through the Internet. For example, distributed file storage, distributed file access, and video streaming applications in the community share network resources in the WMNs without using the Internet, which improves the performance of these applications.
Neighborhood community networks allow faster and easier dissemination of cached information that is relevant to the local community. Mesh routers can be easily mounted on rooftops or windows and the client devices get connected to them in a single hop. Wireless mesh network-based Community Networking 4.
During disasters like fire, flood, and earthquake, all the existing communication infrastructures might be collapsed. So during the rescue operation, mesh routers can be placed at the rescue team vehicle and different locations which form the high-bandwidth mesh backbone network, as shown in Fig 7.
By providing different communication interfaces at the mesh routers, different mobile devices get access to the network. This helps people to communicate with others when they are in critical situations.
These networks can be established in less time, which makes the rescue operation more effective. Fig 7. In this section, we present different research challenges facing mesh networks. In the context of WMNs, the issues which affect their performance have been characterized below: The aforementioned requirements make the design of the MAC functions highly challenging.
Several distributed multi-hop MAC protocols have been proposed which improve the throughput in multi-hop paths. However they are still far from being optimum solutions to be exploited by the network operator for real commercial breakthrough.
Apart from these, one needs to properly identify the issues related to the spectral efficiency of both high frequency and low frequency mesh systems. Proper characterization for the mesh capacity constraints and the understanding of the used network and its application is very important in determining the practical utility of mesh networks and the enabling technologies.
It is also important to consider how mesh networks could live alongside existing radio systems, in terms of interference and coexistence strategies.
This functionality should be assured by the mesh routing protocol. Some efforts have been initiated to adapt the ad-hoc routing protocols for WMNs. But due to the fact that the ad-hoc routing protocols lack various important performance factors such as scalability, fault tolerance, QOS metrics, load balancing, lack of cross layer interaction etc.
In contrast, certain areas such as mobility and power management in ad-hoc networks and WMNs have totally different requirements, which made ad-hoc routing solutions not appropriate for WMNs.
In order to resolve the above issues, innovative solutions are indispensable in WMNs. Due to the distributed multi hop features of mesh networks and the non-significant support from the lower layers to assure certain quality of service support for the application layer, there is serious need to adapt the existing applications to WMN architecture.
Moreover, considerable efforts are also required to discover innovative new applications which at first, make the life of the customer easier and craft the WMNs to appear as a distinctive wireless network solution, at second.
WMN is a potential candidate technology to enable the integration of various existing networks through gateway functionalities in the mesh routers. However, active research should be performed in this interesting domain to enable the users to seamlessly utilize services irrespective of the network concerned, which is actually providing that service. In typical architectures of mesh networks, these factors generally go against each other due to its self-adjusting and non-hierarchical nature.
Scalability problems are even more critical in mobile mesh architectures. When the issue of network responsiveness is added on top of all these characteristics, the equation becomes even more complicated and the guarantees of Quality-of-Service for the clients are disturbed.
The typical scalability issues in multi- hop networking apply for mesh networks too, since multi-hop communications is common in WMN, i. Designing a scalable mesh network requires the proper understanding of the complex inter-relationship between the contrasting network characteristics.
This is especially true for applications that need to handle continuous data streams where high capacity is critical to maintain the scalability and reliability of the network. Careful design and proper characterization of the physical layer mechanisms depending on the envisaged application scenarios and an inherent foresight on the number of users, designing efficient and distributed backbone communication topologies  using hybrid multiple access schemes exploiting the availability of multi-radio, multi-channel systems, devising efficient routing protocols for transporting data robustly etc.
To assure service availability and continuity, Inter domain accounting is important in WMNs. The economic interests require the application of usage sensitive billing systems based on the gathered accounting information for each client. It is recommended that these systems allow online payment or pre-paid tokens. However, processing delay constraints should be considered as well as the need for authentication and integrity. The IEEE The challenge is in providing lightweight implementations for mesh networking techniques considering the limited resources in the digital devices.
Facing the throughput degradation and unfairness in IEEE On the other hand, IEEE The Wi MAX forum is working to ensure the interoperability of manufactured equipments using these standard suites.
IEEE Consequently, In addition, IEEE Its target is to conceive a system for combined fixed and mobile broadband wireless access, operating in the GHz licensed bands. Simultaneously, IEEE To date, the ZigBee standard is the only market-ready wireless mesh standard. Mesh networking is one of the architecture examples defined by this WG and is classified as distributed WLAN architectures .
This WG is looking for extensibility for future applicability to other access technologies especially the Finally, Software Defined Radio SDR benefits from today's high processing power to develop multi-band, multi-standard base stations and terminals .