ad hoc mobile wireless network routing protocols

TECHNIA – International Journal of Computing Science and Communication Technologies, VOL. 2, NO. 1, July 2009. (ISSN 0974-3375)

294

Ad hoc Mobile Wireless Network Routing Protocols-A Review

Ruchi TuliGitarattan International Business School, PSP 2A & 2B, Madhuban Chowk, Rohini, New Delhi

Abstract:- Mobile ad hoc networks(MANET) represent complexdistributed systems that comprise wireless mobile nodes that canfreely and dynamically self organize into arbitrary andtemporary ad hoc network topologies, allowing people anddevices to seamlessly internet work in areas with no preexistingcommunication infrastructure e.g., disaster recoveryenvironments. An ad-hoc network is not a new one, having beenaround in various forms for over 20 years. Traditionally, tacticalnetworks have been the only communication networkingapplication that followed the ad-hoc paradigm. Recently theintroduction of new technologies such as Bluetooth, IEEE 802.11and hyperLan are helping enable eventual commercial MANETdeployments outside the military domain. These recentrevolutions have been generating a renewed and growing interestin the research and development of MANET. To facilitatecommunication within the network a routing protocol is used todiscover routes between nodes. The goal of the routing protocol isto have an efficient route establishment between a pair of nodes,so that messages can be delivered in a timely manner. Routingprotocols used in wired network cannot be used for mobile ad-hoc networks because of node mobility. The ad-hoc routingprotocols are divided into two classes: Table Driven and Demandbased. This paper reviews and discusses routing protocolbelonging to each category, makes a general comparison of tabledriven and on-demand protocols and finally discusses theapplications and challenges faced by ad hoc mobile wirelessnetworks.

Keywords: MANET, Routing Protocol, Table driven, On-demand,active and reactive protocols.

1. INTRODUCTIONWireless network has become very popular in the computingindustry. Wireless network are adapted to enable mobility.There are two variations of mobile network. The first is infra-structured network (i.e. a network with fixed and wiredgateways). The bridges of the network are known as basestations. A mobile unit within the network connects to andcommunicates with the nearest base station (i.e. within thecommunication radius).Application of this network includes office WLAN. Thesecond type of network is infrastructure less mobile networkcommonly known as AD-HOC network. They have no fixedrouters. All nodes are capable of moving and be connected inan arbitrary manner. These nodes function as routers, whichdiscover and maintain routes to other nodes in the network.Ad-hoc networks can be used in areas where there is little orno communication infrastructure or the existing infrastructure

is expensive or inconvenient to use. Some applications of ad-hoc network are students using laptop to participate in aninteractive lecture, business associates sharing informationduring a meeting, soldiers relaying information about situationawareness in a battlefield, and emerging disaster relief after anearthquake or hurricane.An ad-hoc network is a collection of mobile nodes, whichforms a temporary network without the aid of centralizedadministration or standard support services regularly availableon conventional networks. The nodes are free to moverandomly and organize themselves arbitrarily; thus thenetwork’s wireless topology may change rapidly andunpredictably.

2. TAXONOMY OF AD-HOC NETWORKMobile ad hoc network protocol (MANET) routing protocolsare classified according to several criteria, reflectingfundamental design and implementation choices. SimpleMANET architecture is shown in figure 1.

FIG. 1 : SIMPLE MANET ARCHITECTURE

2.1 Structure : Routing protocols are categorized as uniformor non-uniform protocols

Uniform protocols: In a uniform protocol, none of thenodes take a distinguished role in routing scheme; eachnode sends and responds to routing control messages insame way. No Hierarchical structure is imposed in thenetwork.Non-uniform protocols: In this type of protocol routingcomplexity can be limited by reducing the number ofnodes participating in routing computation.

Tuli: Ad hoc Mobile Wireless Network Routing Protocols – A Review

295

A preliminary classification of the routing protocols can bedone via the type of cast property, i.e., whether they use a [1].

Unicast,Geocast,Multicast, orBroadcast forwarding

Broadcast is the basic mode of operation over a wirelesschannel; each message transmitted on a wireless channel isgenerally received by all neighbors located within one-hopfrom the sender. The simplest implementation of the broadcastoperation to all network nodes is by naive flooding, but thismay cause the broadcast storm problem due to redundantrebroadcast. Surveys [2] existing methods for flooding awireless network intelligently. Unicast forwarding means aone-to-one communication, i.e., one source transmits datapackets to a single destination. This is the largest class ofrouting protocols found in ad hoc networks. Multicast routingprotocols come into play when a node needs to send the samemessage, or stream of data, to multiple destinations. Geocastforwarding is a special case of multicast that is used to deliverdata packets to a group of nodes situated inside a specifiedgeographical area. Nodes may join or leave a multicast groupas desired; on the other hand, nodes can join or leave a geocastgroup only by entering or leaving the correspondinggeographical region.

2.2 Scheduling: MANET routing protocols are typicallysubdivided into two main categories [3]:

Proactive routing protocols andReactive on-demand routing protocols

Proactive routing protocols are derived from legacy Internetdistance-vector and link-state protocols. They attempt tomaintain consistent and updated routing information for everypair of network nodes by propagating, proactively, routeupdates at fixed time intervals. As the routing information isusually maintained in tables, these protocols are sometimesreferred to as Table-Driven protocols. Reactive on demandrouting protocols, on the other hand, establish the route to adestination only when there is a demand for it. The sourcenode through the route discovery process usually initiates theroute requested. Once a route has been established, it ismaintained until either the destination becomes inaccessible(along every path from the source), or until the route is nolonger used, or expired [4]. Design constraints that are specificto ad-hoc networking [5] are :

2.2.1 Autonomous and infrastuctureless: MANET does notdepend on any established infrastructure or centralizedadministration. Each node operates in distributed peer-to peermode, acts as an independent router and generates independentdata.

2.2.2 Multi-hop routing: No default router available, everynode acts as a router and forwards each other’s packets toenable information sharing between mobile hosts.

2.2.3 Dynamically changing network topologies: In mobilead-hoc networks, because nodes can move arbitrarily, thenetwork topology which is multi hop can change frequentlyand unpredictably resulting in route changes, frequent networkpartitions and possibly packet losses.

2.2.4 Variation in link and node capabilities: Each node isequipped with one or more radio interfaces that has varyingtransmission/receiving capabilities and operate across differentfrequency bands [6,7]. This heterogeneity in node radiocapabilities can result in possibly asymmetric links. Each nodehas a different software/hardware configuration resulting invariability in processing capabilities.

2.2.5 Energy constrained operation: Because batteriescarried by each mobile node have limited power supply so theprocessing power is limited, which in turn limits the servicesand application that can be supported by each node. Thisbecomes a bigger issue in MANET because each node acts asboth an end system and a router at the same time; additionalenergy is required to forward packets from other nodes.

2.2.6 Network scalability: Many MANET applicationsinvolve large networks with tens of thousands of mobile nodesas found for example in sensor and tactical networks [8].Scalability is critical to the successful deployment of thesenetworks.

3. ROUTING PROTOCOLSAs shown in Figure 2 routing protocols may be generallycategorized as

1. Table driven2. Source initiated (demand driven)

3.1 Table Driven Protocols: Table driven protocols maintainconsistent and up to date routing information about each nodein the network. These protocols require each node to storetheir routing information and when there is a change innetwork topology updation has to be made throughout thenetwork. The different types of Table driven protocols are:

Destination sequenced Distance vector routing (DSDV)Wireless routing protocol (WRP)Fish eye State Routing protocol (FSR)Optimised Link State Routing protocol (OLSR)Cluster Gateway switch routing protocol (CGSR)Topology Dissemination Based on Reverse path forwarding(TBRPF)


296

The Destination-Sequenced Distance-Vector (DSDV) protocol[8] is a distance-vector protocol with extensions to make itsuitable to MANET. Every node maintains a routing tablewith one route entry for each destination in which the shortestpath route (based on number of hops) is recorded. To avoidrouting loops, a destination sequence number is used. A nodeincrements its sequence number whenever a change occurs inits neighborhood. CGSR extends DSDV with clustering toincrease the protocol scalability [9]. WRP is another loop-freeproactive protocol where four tables are used to maintaindistance, link cost, routes and message retransmissioninformation [10]. Loop avoidance is based on providing forthe shortest path to each destination both the distance and thesecond-to-last hop (predecessor) information. Despite thevariance in the number of routing tables used, and thedifference in routing information maintained in these tables,proactive routing protocols like DSDV, CGSR and WRP areall distance vector shortest-path based, and have the samedegree of complexity during link failures and additions.

OLSR protocol [11] is an optimization for MANET of legacylink-state protocols. The key point of the optimization is themultipoint relay (MPR). Each node identifies (among itsneighbors) its MPRs. By flooding a message to its MPRs, anode is guaranteed that the message, when retransmitted bythe MPRs, will be received by all its two-hop neighbors. LikeOLSR, TBRPF [12] is a link-state routing protocol thatemploys a different overhead reduction technique. Each nodecomputes a shortest path tree to all other nodes, but tooptimize bandwidth only part of the tree is propagated to theneighbors. The FSR protocol [13] is also an optimization overlink-state algorithms using fisheye technique. LANMAR [14]builds on top of FSR and achieves hierarchical routing bypartitioning the network nodes into different mobility groups;a landmark node is elected within each group to keep track ofwhich logical subnet a node belongs to, and facilitate inter-group routing; FSR is used for intragroup routing.

FIG. 2 CATEGORIZATION OF AD HOC ROUTING PROTOCOLS

3.2 Source Initiated Demand driven: In on-demand routingprotocols routes are created as and when required. When asource wants to send to a destination, it invokes the routediscovery mechanisms to find the path to the destinations. Theroute remains valid till the destination is reachable or until theroute is no longer needed.

The different types of On Demand driven protocols are:

Ad hoc On Demand Distance Vector (AODV)Dynamic Source Routing protocol (DSR)Temporally ordered routing algorithm (TORA)Associativity Based routing (ABR)

These protocols depart from the legacy Internet approach. Toreduce the overhead, the route between two nodes isdiscovered only when it is needed. Representative reactiverouting protocols include: Dynamic Source Routing (DSR),Ad hoc On Demand Distance Vector (AODV), TemporallyOrdered Routing Algorithm (TORA), Associativity BasedRouting (ABR), and Signal Stability Routing (SSR).

DSR is a loop-free, source based, on demand routing protocol[15], where each node maintains a route cache that containsthe source routes learned by the node. The route discoveryprocess is only initiated when a source node do not alreadyhave a valid route to the destination in its route cache; entriesin the route cache are continually updated as new routes arelearned. AODV is a reactive improvement of the DSDVprotocol. AODV minimizes the number of route broadcasts bycreating routes on-demand [16], as opposed to maintaining acomplete list of routes as in the DSDV algorithm. Similar toDSR, route discovery is initiated on-demand, the route requestis then forward by the source to the neighbors, and so on, untileither the destination or an intermediate node with a freshroute to the destination, are located. In general, both AODVand DSR work well in small to medium size networks withmoderate mobility.

TORA is another source-initiated on-demand routing protocolbuilt on the concept of link reversal of the Directed AcyclicGraph (ACG) [17]. In addition to being loop-free andbandwidth efficient, TORA has the property of being highlyadaptive and quick in route repair during link failure, whileproviding multiple routes for any desired source/destinationpair. These properties make it especially suitable for large,highly dynamic, mobile ad hoc environments with densenodes populations. ABR protocol is also a loop free protocol,but it uses a new routing metric termed degree of associationstability in selecting routes, so that route discovered can belonger-lived route, thus more stable and requiring less updatessubsequently. Signal Stability Algorithm (SSA) [18] isbasically an ABR protocol with the additional property ofroutes selection using the signal strength of the link.


297

4. COMPARISON OF ON-DEMAND PROTOCOLS ANDTABLE-DRIVEN ROUTING PROTOCOLS

Parameters On-Demand Table-Driven

Availability of routinginformation

Availability whenneeded

Always availableregardless of need

Routing philosophy Flat Mostly flat except forCGSR

Period route updates Not required RequiredCoping with mobility Use localized route

discovery as in ABRand SSR

Inform other nodes toachieve a constantrouting table

Signaling trafficgenerated

Grows with increasingmobility of activeroutes (as in ABR)

Greater than that ofon-demand routing

Quality of servicesupport

Few can support QoS,although most supportshortest path

Mainly shortest pathas the Qos metric

Table 1. Overall comparisons of on-demand versus table-driven routingprotocols.

As discussed earlier, the table-driven ad hoc routing approachis similar to the connectionless approach of forwardingpackets, with no regard to when and how frequently suchroutes are desired. It relies on an underlying routing tableupdate mechanism that involves the constant propagation ofrouting information, this is not the case, however, for on-demand routing protocols. When a node using an on-demandprotocol desires a route to a new destination, it will have towait until such a route can be discovered. On the other hand,because routing information is constantly propagated andmaintained in table-driven routing protocols, a route to everyother node in the ad hoc network is always available,regardless of whether or not it is needed. Table 1 lists some ofthe basic differences between the two classes of algorithms.Another consideration is whether a flat or hierarchicaladdressing scheme should be used. All of the protocolsconsidered here, except for CGSR, use a flat addressingscheme. In [19] a discussion of the two addressing schemes ispresented. While flat addressing may be less complicated andeasier to use, there are doubts as to its scalability.

5. RESULTSIn general, on-demand reactive protocols are more efficientthan proactive ones. On-demand protocols minimize controloverhead and power consumption since routes are onlyestablished when required. By contrast, proactive protocolsrequire periodic route updates to keep information current andconsistent; in addition, maintain multiple routes that mightnever be needed, adding unnecessary routing overheads.Proactive routing protocols provide better quality of servicethan on-demand protocols. As routing information isconstantly updated in the proactive protocols, routes to everydestination are always available and up-to-date, and hence

end- to- end delay can be minimized. For on-demandprotocols, the source node has to wait for the route to bediscovered before communication can happen.

In addition to proactive and reactive protocols, another classof unicast routing protocols that can be identified is that ofhybrid protocols. The Zone- Based Hierarchical Link StateRouting Protocol (ZRP) is an example of hybrid protocol thatcombines both proactive and reactive approaches thus tryingto bring together the advantages of the two approaches. ZRPdefines around each node a zone that contains the neighborswithin a given number of hops from the node.In terms of metrics comparisons between the two routingprotocols are: Throughput: proactive protocols perform wellthen reactive protocol; End to end delay: proactive protocolsperform well than reactive protocols; Routing load: reactiveprotocols perform well than proactive protocols.

6. APPLICATIONS AND CHALLENGESAkin to packet radio networks, ad hoc wireless networks havean important role to play in military applications. Soldiersequipped with multimode mobile communicators can nowcommunicate in an ad hoc manner without the need for fixedwireless base stations. In addition, small vehicular devicesequipped with audio sensors and cameras can be deployed attargeted regions to collect important location andenvironmental information which will be communicated backto a processing node via ad hoc mobile communications.Commercial scenarios for ad hoc wireless networks include:

Conferences/meetings/lecturesEmergency servicesMilitary environmentPersonal Area Networking

People today attend meetings and conferences with theirlaptops, palmtops, and notebooks. It is therefore attractive tohave instant network formation, in addition to file andinformation sharing without the presence of fixed base stationsand systems administrators. Ad hoc mobile communication isparticularly useful in relaying information (status, situationawareness, etc.) via data, video, and/or voice from one rescueteam member to another over a small handheld or wearablewireless device. Current challenges for ad hoc wirelessnetworks include:

MulticastQoS supportPower-aware routingLocation-aided routing

As mentioned above, multicast is desirable to supportmultiparty wireless communications. Since the multicast treeis no longer static (i.e., its topology is subject to change overtime), the multicast routing protocol must be able to cope with


298

mobility, including multicast membership dynamics (e.g.,leave and join). In terms of QoS, it is inadequate to considerQoS merely at the network level without considering theunderlying media access control layer. Another importantfactor is the limited power supply in handheld devices, whichcan seriously prohibit packet forwarding in an ad hoc mobileenvironment. Hence, routing traffic based on nodes’ powermetrics is one way to distinguish routes that are more long-lived than others. Finally, instead of using beaconing orbroadcast search, location- aided routing uses positioninginformation to define associated regions so that the routing isspatially oriented and limited.

7. DISCUSSIONIn coming years, mobile computing will keep flourishing, andan eventual seamless integration of MANET with otherwireless networks, and the fixed Internet infrastructure,appears inevitable. Ad hoc networking is at the center of theevolution towards the 4th generation wireless technology. Theopportunity and importance of ad hoc networks is beingincreasingly recognized by both the research and industrycommunity, as evidenced by the flood of research activities, aswell as the almost exponential growth in the Wireless LANsand Bluetooth sectors.In moving forward towards fulfilling the opportunity, thesuccessful addressing of open technical and economical issueswill play a critical role in achieving the eventual success andpotential of MANET technology. Despite the large volume ofresearch activities and rapid progress made in the MANETtechnologies in the past few years, almost all research areas(from enabling technologies to applications) still harbor manyopen issues. There has been good progress in studying theprotocols behavior. The perception is that large number ofcompeting routing protocols lack of WGwide consensus andfew signs of convergence. To overcome this situation, adiscussion is currently ongoing to focalize the activities of theMANET Working Group (WG) towards the design of IETFMANET standard protocols and to split off related long termresearch work from IETF. MANET WG proposes a view ofmobile ad hoc networks as an evolution of the Internet. Thismainly implies an IP-centric view of the network, and the useof a layered architecture. Current research points out thoughthat this choice may limit developing efficient solutions forMANET. Other promising directions have been identified[20]. The use of the IP protocol has two main advantages: itsimplifies MANET interconnection to the Internet, andguarantees the independence from wireless technologies. Onthe other hand, more efficient and lightweight solutions can beobtained, for example, by implementing routing solutions atlower layers [21]. The layered paradigm has highly simplifiedInternet design, however when applied to ad hoc networks, itmay result in poor performance as it prevents exploitingimportant interlayer dependencies in designing efficient adhoc network functions. Relaxing the Internet layeredarchitecture, by removing the strict layer boundaries, is an

open issue in the MANET evolution. Cross-layer design ofMANET architecture and protocols is a promising directionfor meeting the emerging application requirements,particularly when energy is a limited resource.

8. CONCLUSIONIn this paper description of several routing schemes proposedfor ad-hoc mobile networks has been provided. Aclassification of these schemes according the routing strategy.i.e., table driven and on demand has also been listed. Acomparison of these two categories of routing protocols,highlighting their features, differences and characteristics isdiscussed. Finally, possible applications and challenges facingad-hoc wireless networks have also been identified.

While it is not clear that any particular algorithm or class ofalgorithm is the best for all scenarios, each protocol hasdefinite advantages and disadvantages and has certainsituations for which it is well suited. The field of ad-hocmobile networks is rapidly growing and challenging, andwhile there are still many challenges that need to be met, it islikely that such networks will see wide-spread use within thenext few years.

REFERENCES

1. Kuosmanen, P., 2002. Classification of ad hoc routing protocols, FinnishDefence Forces, Naval Academy, Finland. Available from<http://keskus.hut.fi/opetus/s38030/k02/Papers/12-Petteri.pdf>.

2. Stojmenovic, I., and J. Wu, 2003. Broadcasting and activity-schedulingin ad hoc networks, in: S. Basagni, M. Conti, S. Giordano, I.Stojmenovic (Eds.), Ad Hoc Networking, IEEE Press Wiley, New York.

3. Belding-Royer, E.M., and C.-K. Toh, 1999. A review of current routingprotocols for ad-hoc mobile wireless networks, IEEE PersonalCommunications Magazine, 46–55.

4. Freebersyser, J.A., and Barry Leiner, 2001. A DoD perspective onmobile ad hoc networks, in: Charles E. Perkins (Ed.), Ad HocNetworking, Addison Wesley, Reading, MA, pp. 29–51.

5. Corson, M.S., J.P. Maker, and J.H. Cernicione, 1999. Internet-basedmobile ad hoc networking, IEEE Internet Computing 3 (4) pp. 63–70.

6. Chlamtac, I and A. Lerner, 1986. Link allocation in mobile radionetworks with noisy channel, in: IEEE INFOCOM, Bar Harbour, FL.

7. Chlamtac, I, and A. Lerner, 1987. Fair algorithms for maximal linkactiviation in multi-hop radio networks, IEEE Transactions onCommunications COM-35 (7) pp 739-746.

8. Perkins, C.E., and P. Bhagwat, 1994. Highly dynamic destinationsequenced distance-vector routing (DSDV) for mobile computers,Computer Communications Review pp. 234–244.

9. Chiang, C.C., H.K. Wu, W. Liu, and M. Gerla, 1997. Routing inclustered multihop, mobile wireless networks with fading channel, in:Proceedings of IEEE SICON-97, pp. 197–211.

10. Murthy, S., and J.J. Garcia-Luna-Aceves, 1996. An efficient routingprotocol for wireless networks, ACM Mobile Networks andApplications (MONET) Journal, Special Issue on Routing in MobileCommunication Networks, pp. 183–197.

11. Jacquet, P, P. Muhlethaler, and A. Qayyum, 1998. Optimized Link StateRouting Protocol, Internet Draft, draft-ietf-manetolsr- 00.txt.


299

12. Bellur, B, R.G. Ogier, and F.L. Templin, 2001. Topology broadcastbased on reverse-path forwarding (TBRPF), IETF Internet Draft,draftietf- manet-tbrpf-01.txt.

13. Pei, G., M. Gerla, and T.-W. Chen, 2000. Fisheye state routing in mobilead hoc networks, in: Proceedings of the 2000 ICDCS Workshops,Taipei, Taiwan, pp. D71–D78.

14. Pei, G., M. Gerla, and X. Hong, 2000. LANMAR: landmark routing forlarge scale wireless ad hoc networks with group mobility, in:Proceedings of IEEE/ACM MobiHOC 2000, Boston, MA, pp. 11–18.

15. Johnson, D.B., and D.A. Maltz, 1996. Dynamic source routing in adhocwireless networks, in: T.Imielinski, H. Korth (Eds.), Mobile Computing,Kluwer Academic Publishers, pp. 153–181.

16. Perkins, C.E., and E.M. Royer, 1999. Ad-hoc ondemand distance vectorrouting, in: Proceedings of 2nd IEEE Workshop on Mobile ComputingSystems and Applications, pp. 90-100.

17. Park, V.D., and M.S. Corson, 1997. A highly adaptive distributedrouting algorithm for mobile wireless networks, in: Proceedings ofINFOCOM- 97 Volume: 3, pp. 1405-1413.

18. Dube. R., et. al., 1997. Signal Stability based Adaptive Routing (SSA)for Ad Hoc Mobile Networks, IEEE Pers., Communication, pp. 36-45.

19. D. Baker et al., “Flat vs. Hierarchical Network Control Architecture,”ARO/DARPA Wksp. Mobile Ad-Hoc Networking; http:llwww.isr.umd.edu/Courses’Workshops/MANET/program.html. Mar. 1997.

20. Goldsmith, A.J., and S.B. Wicker, 2002. Design challenges for energy-constrained ad hoc wireless networks, IEEE Wireless Communications 9(4) pp.8-27.

21. Acharya, A., A.Misra, and S.Bensal, 2002. A labelswitching packetforwarding architecture for multihop wireless LANs, in: M.Conti,D.Raychaudhuri(Eds), Proceedings of the ACM Workshop on MobileMultimedia (WoWMoM 2002), Atlanta, GA.

ad hoc mobile wireless network routing protocols

Documents