Summary of wireless multi-hop mesh networks
The technology of wireless networks is developing fast, and the applications of wireless networks offer the publics and society much convenien
ce. But, at present the traditional wireless net is just used to replace the cable simply. Ⅰ. Traditional WLAN Along with the expanding of the wireless net’s scale, the requiring of wired connection for each access-point in WLAN makes it meet a lot of challenges and in convenience in the environment of lack of cable infrastructure. The traditional WLAN shows it’s insufficient gradually. The disadvantages of traditional WLAN: a. The poor reliability: In the traditional WLAN, several users get access to the wireless net through one access-point directly—it is call “single-hop” networks. So long as one stoppage can breakdown the whole networks.
Figure1. Traditional WLAN b. The small coverage: The normal technology of “point-to-point” or “point-to-multipoint” uses the short networks, with the limitations of coverage, just like networks of 802.11 and Bluetooth. c. The poor scalability: As the joining and sharing data flow of access-points of extra nodes, the bandwidth of whole networks will come down gradually. d. The bad quality of communication: Because of noise in link and mistakes in communication, the bandwidth will decrease as the distance’s increasing. Most kind of wireless networks have “blind points” in its effective distance,
in which it shows the poor signal quality. Ⅱ. The concept of wireless mesh network (WMN) With the evolution of communication technique and catering to the need of the market, a new wireless network emerges, which is called “wireless mesh network”, short for “WMN”. Wireless mesh network is a flexible communication system to improve the efficiency and quality in single transmission. WMNs have emerged as a cost-effective alternative for large-area communications. WMNs have also become an important technology for constructing wireless networks beyond the 3G high-speed cell phone networks. A mesh network is an ad hoc network that employs one of two connection arrangements: full mesh topology or partial mesh topology. WMNs do not require wired infrastructure for long-haul communications, thus offering the major benefit of low-cost and flexible deployment and maintenance. WMNs, however, suffer from the same security vulnerabilities as conventional wireless networks. Because data are transmitted in the form of radio waves in the air without physical boundaries, packet eavesdropping and packet intercepting can be carried out easily by attackers with the help of relatively cheap equipment. Moreover, wireless channels may be disturbed by noise and interference, which affect the quality of communications. WMNs are multi-hop networks, which are especially vulnerable to interference problems: Adjacent hops on the same path and neighboring paths could interference with each other. WMNs are distributed networks, which have additional issues on routing, authentication, and key management. Therefore, security is a major concern in WMNs. Ⅲ. Ad hoc networks and Wireless mesh networks
Figure2. Ad hoc network a. Ad hoc networks Ad hoc networks are wireless networks that can be deployed quickly. A mobile ad hoc
network is a potential solution for deploying a data and voice network for FRs in scenarios with limited or no infrastructure. All nodes in the network act as equal peers as Figure 2. An ad hoc network typically uses a suitable routing protocol such as AODV or DSR to allow multi-hop communication between the nodes. One or more nodes may act as gateways and connect the ad hoc network with the Internet or some other network. The gateways may use a wired or even a wireless back-haul to exchange data with an external network. Self-contained ad hoc networks may not even have a back-haul. Ad hoc networks generally have only single-radio nodes and can be considered to be single-tier mesh networks. b. Wireless mesh networks In the full mesh topology, each node is connected directly to each other node. In the partial mesh topology, nodes are connected to only some, but not all the other nodes. Wireless mesh networks thus combine a mesh topology with ad hoc wireless network characteristics. Mesh networks can be built on a mix of ?xed and mobile nodes interconnected via wireless links to form a multi-hop ad hoc network.
Figure3. Wireless multi-hop mesh network of two tiers In a sufficiently dense wireless mesh network, there is rarely a single point of failure and the network is able to recover from the failure of an individual node by routing around it to other nodes in a direct or multi-hop manner. A mesh network allows nodes to communicate with other nodes without being routed through a central switch point, thus eliminating centralized failure and providing self-healing and self-organization behavior. As shown in Figure 3, several mesh routers act as APs to the mobile clients. The mesh provides a wireless infrastructure to mobile clients using an ad hoc wireless back-bone to route packets to their respective destinations. One or more
mesh routers can act as gateways for the back-haul traffic to external networks, and may use a wired or wireless connection. An ad hoc wireless network protocol, such as Ad hoc On-demand Distance Vector routing, Optimized Link State Routing, or Dynamic Source Routing, is typically used for routing packets in the network between mesh routers. There exist over 70 po-tential protocols that could be used for routing in a mesh network. Path diversity and redundancy are the key features of a successful mesh network. Two-tier mesh networks, such as that shown in Figure3, avoid the completely self-organized nature of ad hoc networks. Such networks have two or more radios. Typically one radio is used in the “access tier” as an AP to provide wireless service to mobile clients in the area connected to it (just like in a Wi-Fi access point). The second radio is used in the “mesh tier” by the mesh nodes to form a wireless backbone, using a suitable mesh routing protocol. In this context, the term “mesh routing protocol” is synonymous with the term “ad hoc routing protocol”. Thus, each mesh node in a two-tier wireless mesh network provides the functionality of both an access point as well as router. The back-haul of the mesh node wirelessly relays the traffic from mesh router to mesh router until it reaches a gateway. The gateway node then connects to the Internet or to another private network (through a wired connection). Another option is to use an alternative technology such as satellite or cellular network links, to wirelessly connect the mesh network with an external network. We use this two-tier model in Dart-Mesh, our mesh implementation testbed. One bene?t of this architecture is that for end users the experience of joining such a network is identical to the process for a client joining a Wi-Fi infrastructure-based WLAN. This process is well supported by all compatible wireless client devices that support the published standards of the Wi-Fi device manufacturers consortium. The clients do not need to run any special routing software or con?gure the radio to run in “ad hoc mode.” Indeed, the “ad hoc mode” is not fully supported by many device manufacturers, device drivers and operating systems, and con?guring a wireless card to use this mode is not an easy process. c. Relationship between ad hoc networks and Wireless mesh networks Ad-hoc networks and WMNs are often used interchangeably, but there are subtle differences between them on a number of aspects. First, end-user devices in ad-hoc networks also perform routing and configuration functionalities for other nodes. In WMNs these functionalities are performed by mesh routers. In terms of functionalities, mesh client devices are the same as those in IEEE 902.11 WLANs, providing lower energy consumption and high-end application capabilities to mobile users and energy constrained users. Moreover, because end-users do not support routing functionalities the cost of WMN client devices may be reduced. Second, WMNs adopt the multiple-radio interface technology, which differs from ad-hoc networks. The multi-radio interface technology makes it possible to separate routing (control packets) from data transmission (data packets). This significantly improves the capacity of the network. On the other hand, routing and data transmission in ad-hoc networks are performed on the same radio interface, which would reduce the performance. Third, because end-user devices in ad-hoc networks provide routing functionalities, the network topology and connectivity depend on the movement of users. This imposes additional challenges to routing protocols, network configurations, and network deployment. Different from
ah-hoc networks, routing is performed by mesh routers that are either stationary or slightly mobile. Mesh routers constitute the infrastructure for WMNs. While providing continuous connectivity throughout the networks, WMNs continue to support mobile end-users without compromising the performance. The differences between ad-hoc networks and WMNs are subtle, making it difficult to draw a clean line between the two types of networks. It may be helpful to view ad-hoc networks as a subset of WMNs or vice versa. Ⅳ. Challenges and problems to be solved in the research of Wireless mesh networks Several challenges need to be overcome to realize a wireless mesh network suitable for First Responders. The overarching challenge is to engineer and build a wireless mesh network that provides all the desired characteristics in a cost effective manner. We enumerate some of the desirable characteristics of a mobile mesh network and their corresponding challenges: a. Robustness: The mesh network should be functional even if a few individual nodes fail. For example, in a hazardous scenario a ?re may wipe out some sensors and nodes. Thus mesh nodes should be capable of operating in extreme conditions and have sufficient battery life. Mesh nodes which are heat and environment resistant are likely to be expensive to manufacture. Larger and heavier nodes could reduce mobility, making deployment harder. b. Reliability: The network must maintain connectivity between the majorities of critical nodes during the operation in spite of the potential failure of individual nodes. Most client nodes have limited range and limited power, so depending on the scenario we may wish to optimize for a network that functions in a limited manner for a long time or one in which all nodes run for a shorter time with full power and maximum connectivity. Node mobility can cause links to break. Environmental conditions and physical obstructions can cause temporary or permanent link interruptions. c. Service quality: The network should provide enough bandwidth to meet the needs of all users or at least high-priority users. The traf?c patterns of data in this network will depend on the characteristics of the individual applications that run on each client. d. Inter-operability: Communication must be secure yet allow different agencies to communicate with each other when operations require interdepartmental cooperation. e. Con?guration: The users would like the network to be quick to set up (almost instantaneous) with preferably zero-on-site con?guration. The solution must be easy to deploy. For example, the network manager must ensure that all nodes have the necessary encryption keys and that all nodes are con?gured correctly with the right settings. f. Management: We should be able to maintain connectivity and desired network performance. We need to monitor the mesh network to become aware of any problems with the nodes, links or the routing of traf?c. We need to identify potential problem areas, nodes that may need assistance or replacement, and help the administrator choose the right corrective action. For example, a node that is about to run out of power should be identi?ed, so someone can manually recharge it or to inform other nodes, to allow them to ?nd alternate routes around the weaker node.
Monitoring the status of many nodes in a scalable manner is a challenge. Management traf?c should not affect the network traf?c yet be frequent enough to be meaningful and adaptive to changing network conditions. g. Location Information: Awareness of the location of each FR and each network device is useful for managing the situation on-site. Location can be determined by Global Positioning Systems (GPS) outdoors and by localization of signal strength of neighboring nodes and wireless routers (indoors). Current GPS technologies do not work well indoors or underground. h. Security: Depending on the nature of the deployment, privacy, con?dentiality and integrity requirements of transmitted data must be ensured. Robustness against jamming and attacks or malicious users is desirable. It may be impossible to guarantee resilience in the face of physical layer jamming and other possible attacks. i. Efficient Routing: Control overhead to route packets in the network should be as low as possible and the routing protocol should provide desired performance. Several mesh routing protocols that use different techniques and optimizations are available. It is impossible to qualitatively compare each protocol to decide which is optimal for a given situation or is the best option for all dynamic situations. j. Updates: It should be possible to update the software on the mobile nodes remotely when required. In a FR scenario updates are likely to occur rarely (if at all) during an event. The mechanism for doing a code update can be complicated. The update operation must be fail-safe and should not disrupt the functionality of other nodes in the network or the network k. Isolation: FRs should not have to share bandwidth with civilians, who may flood the network in an emergency setting. The mesh network should not share the same radio spectrum as civilians, since uncontrolled civilian traffic can flood the shared channel with non-emergency traffic. If FRs use a system isolated from available infrastructure, they may miss out on the potential of using pre-existing infrastructure for routing the network traffic. FR data could be sent at a higher priority or all non-FR users could be disconnected from a pre-existing network (say, a wireless mesh used for community Internet service). V. Wireless mesh networks in application a. Advantages of WMNs in application 1. The low cost of infrastructure: At present, the main method of getting access to Internet is setting Wi-Fi in the hot-points area, and it’s a wireless networks composed of one or several WLANs. Uses in the wireless networks connect Internet through WLAN based on 802.11. Because of the limited coverage of 802.11, it’s necessary to set up a lot of access-points, which connect Internet by cable or optical fiber and cost a lot, to make sure that everywhere in the whole city can get access to wireless nets. In addition, the overmuch cable in use can lower the speed of infrastructure badly. While, as with WMNs, it just need a few access-points to Internet, which can decrease the cost and quicken the course of infrastructure. 2. Admirable reliability of communications: Due to that WMN supply more than one transmission route for every user and routing
algorithm and scheduler program will choose the optimal route for users, which improve the reliability of network and decrease the failures and bottlenecks in communication. Users can choose anyone of these several access-points to connect Internet, so it has a strong ability of error-recovery. 3. Strong ability of self-management: Adopting a “point to point” network, WMN has a distributed system, which has all the advantages of Ad Hoc, just like self-configuration and self-healing. Therefore, it’s automatic and transparent for users to get in networks. If there are new nodes added to WMN, those nodes will search the optimal route to get access to Internet through finding all the wireless routers by the ability of WMN.
b. WMNs in PANs, LANs, and MANs 1. Wireless mesh PANs: Wireless personal area networks (PANs) offer signal ranges in the neighborhood of 1m to 100m, and a wide variety of data rates. Moreover, since the kinds of devices we wish to equip with PANs are often mobile and lightweight, power is at a premium. Thus, the low power consumption of PAN radios is very important to their acceptance. Connecting a number of wireless PANs via multi-hop connection forms a wireless mesh PAN, which helps solve the power constraints of devices in wireless PANs and increase the coverage area and network throughput while reducing transmission power of devices. Comparing with wireless mesh LANs and MANs, wireless mesh PANs are facing certain special challenges. In particular, because an individual PAN covers an area with a diameter of about 10 meters, a mesh PAN may just cover an area with a diameter less than 100 meters, which makes it possible for any node in the mesh PAN to collect the routing information of the entire network in a short period of time. On the other hand, typical PAN devices are power constrained, and so routing protocols for mesh PANs must be simple and efficient. Moreover, stream media applications have become increasingly popular on PANs, which raise an issue of providing QoS in routing protocol is emergent. 2. Wireless mesh MANs: A wireless mesh MAN is a wireless network that covers a geographical territory larger than a building or campus, but smaller than a state or country . Multimedia applications are the main-stream applications on wireless MANs which require high bandwidth, reliable connectivity, and support of free mobility of customer devices. The wireless mesh MANs technology that supports multi-hop communications can meet these requirements, which is more suitable than the conventional point-to-point and point-to-multipoint modes of wireless MANs. Comparing with wireless mesh PANs and LANs, the wireless mesh MANs support communications over wider areas, with many more routing devices and client devices. In addition to the common issues in all three types of wireless mesh networks, wireless mesh WANs must also adopt an efficient scheduling mechanism to cooperate channel assignments. 3. Wireless mesh LANs: A wireless local-area network (LAN) is a wireless network that provides connectivity to wireless devices within a limited area, typically in a building, or a small cluster of buildings. WLANs are widely used, and so connecting multiple WLANs using the mesh technology has attracted much attention. Wireless mesh LANs support multi-hop wireless communications between its APs. This feature offers the benefits of lower deployment cost, flexible scalability,
reliable connection, fault tolerance, load balancing, and optimal resources utilization. Wireless mesh LANs may be implemented on two types of architecture, namely, the infrastructure architecture and the peer-to-peer architecture. The latter is also known as the client meshing architecture, where client devices are directly connected to form a peer-to-peer network and perform routing and configuration functionalities as well as providing end-user applications to customers. It is therefore difficult to draw a clean line between the client meshing architecture and the ad-hoc network architecture. The client meshing architecture of mesh WLANs may sometimes be viewed as the conventional ad-hoc network architecture. In the infrastructure architecture the network is divided into two parts: mesh routers and user devices. The role of mesh routers is similar to that of the APs in traditional infrastructure WLANs, where mesh routers are able to directly communicate with each other through wireless channels. The infrastructure mesh network architecture is scalable and supports gateway functionalities such as bridging, thus enabling connection of WLANs to the Internet and integration with other network technologies. Mesh routers in infrastructure WMNs are not required to be mobile. Therefore, router mobility is no longer a critical challenge in designing routing protocols for wireless mesh LANs. The other challenges of wireless mesh networks, such as QoS guarantees, network management, and various security issues, still remain difficult research problems. Prospect of WMNs in application Compared with the traditional switching networks, WMN overcome the demand of cable between nodes and nodes, which still has the ability of redundancy mechanism and rerouting of distributed networks. In WMNs, if there are some new equipments added, it’s just need to get power for those new equipments, due to those can conduct self-configuration and choose the optimal multi-hoop route. Also the networks can realize the changes of topology and adjust the communication routing automatically to obtain the most effective transmission route. Nowadays, WMNs have a promising prospect in the fields of daily life, commercial and public application. 1. Applications in household: One of important uses of WMN technique is to set up the household wireless network (HWN). The household wireless network can connect PCs, tap tops, HDTV or DVD players and other electronic equipments with out complex wiring and installation. In HWN, every electronic equipment is the user of the network, also is one component of network facility to serve other equipments to get access to the network. The other good use of WMN in household is that it can support highly centralization of bandwidth in application, like HD video. 2. Applications in enterprise: The wireless communication systems of most enterprises adopt the traditional cell phone radio-links, which offer users the “point to point” transmissions. Different from traditional links, WMNs allow users in network share the bandwidth, which can achieve the network load balancing, eliminating the bottleneck of “single-hop” networks. And it is easier and more flexible for adding or adjusting AP in WMNs, which cost less in installation and use. Especially for those enterprises that need to get mobile access to network frequently, the “multi-hoop” structure and flexible configuration of WMNs are to the benefit of topological structure’s adjusting and updating. 3. Applications in campus: The wireless networks of campus are similar with ones in enterprises, but also its own
characteristics. First, the scale of campus WLAN is large, the number of users is great and the communication traffic is also huge, due to compared with users in enterprises the students will use multimedia equipment more frequently; Second, it needs a high rate of coverage. The network must realize seamless roaming indoor, outdoor, in dormitory, in library and other public places; Third, network load balancing is significant especially. Because of students usually get together to anticipate in activities, communication jams may happen sometimes. The traditional method to solve those problems is setting up high-density AP indoor, while setting up few outdoor. On account of that the variation of users’ demand is big, it needs to add new APs or adjust the location of APs, which will bring much cost. As with WMN, it not only is convenient to realize the structure updating of networks, but also can realize the seamless roaming indoor and outdoor. 4. Applications in the hospitals: WMN offers an ideal networking blueprint for the public places like hospitals. Because structures of the buildings in the hospitals are concentrated and complex, hospitals is one of the most difficult places for wireless networks setting. The networks in the hospitals have two main characteristics: First, wiring is difficult. In the traditional wiring, it needs to damage the ensemble of buildings to wiring for the whole networks, which is obviously not helpful for the adjusting of network topology; Second, it needs a very effective robustness for the networks in the hospitals. When there are very important activities held (just like an operation), any stoppage in the networks will bring disastrous consequences. It is an ideal blueprint of adopting WMN to solve those problems. If we want to adjust the network topology, it just needs to move the existing mesh nodes’ locations or set up new mesh nodes, which is very simple and convenient to handle. In addition, the efficient robustness and high bandwidth of WMN make it more appropriate to be applied in such public places as hospitals. 5. Applications in places for journey and leisure: WMN also is appropriate for those distant places those is too difficult or cost too much for wiring of traditional networks, where it still needs to offer wireless Internet service for users, like holiday villages, motels, and remote tourist attractions. WMN can provide wireless Internet service for those sites in a lowest cost. 6. Application in rapid provisioning and temporary adhibition: As with those places need rapid provisioning and temporary adhibitions, like exhibitions, trade fairs, and disaster rescues, WMN undoubtedly is the most economical and effective networking mode. It can provide the superior service in a lower cost. Ⅵ. Conclusion This paper reviewed the development background, development trend, key technology, application prospect and some other hot issues of wireless mesh networks (WMN). Compared with traditional WLAN, WMN has some unparalleled superiorities, such as rapid provisioning and easy installation, NLOS, robustness, flexible structure and high bandwidth. In the WMN networks, one single node not only can pass and receive messages, but also can
retransmit messages to nearby nodes as a router, and the total bandwidth is greatly increasing with the connection of more added nodes and augment of possible routes. In addition, due to the short transmission range of each “short-hop”, the power that transmission need is low. Because of that “multi-hop” usually transmit data to nearby nodes in a lower power with the lower wireless signals interferences between node and node, which makes quality and efficiency of channel signals improved in WMN, it can realize higher network capacities in WMN. For example, in the high-density city networks, WMN can reduce the wireless signals interferences between nearby users, which enhances utilization ratio of channels. However, though the technique of wireless mesh networks has comprehensive application prospects, it also has some problems to hinder its developing, like some issues about interoperability, network delay and network security. After all, WMN is a burgeoning research field in technology of communication, and there are some issues of gordian technique that need to be settled in further research.
REFERENCES 1 Dimitri Bertsekas, Robert Gallager. 2004. Data works. PEARSON EDUCATION ASIA LIMITED and POSTS & TELECOMMUNICATIONS PRESS. 2 BRUNO R, CONTI M, GREGORI E. Mesh networks: Commodity multi-hop Ad Hoc networks [J]. IEEE Com Mag, 2005, 43(3):123-131. 3 Dong Feng, Yan Yongqing. Challenging Problems and Future Applications of wireless mush networks [J]. Jiangsu communication technology, 2006:22-3. 4 Chen Yongjian, Cong Qing. Wireless mesh network and application techniques [J]. Computer Knowledge and Technology, 2010:6-13