Concept Of Multiple Interfaces Between Node Systems Computer Science Essay

In a universe of increasing mobility, there is a increasing demand for people to pass on and hold timely entree to information despite of their location information. A phone call placed from a commuter train may shut a concern trade, remote entree to medical records by a paramedic may salvage a life, or a demand for reconnaissance updates by a soldier with a hand-held device may impact the consequence of a conflict. Each of these cases of nomadic communications pose an technology trial that can be met merely with an efficient, dependable, wireless communicating web. The demand for wireless communicating systems of turning edification and ubiquitousness has led to the demand for a better apprehension of cardinal issues in communicating theory and electromagnetisms and their deductions for the design of highly-competent radio systems.

A radio ad hoc web is a decentralised radio web. The web is ad hoc because each node is willing to send on the information to and from other nodes, and the pick of node that acts as a relay must be made dynamically based on the web connectivity. This is in contrast to the wired webs in which routers perform the undertaking of routing. Similarly in the radio networks a particular node known as an entree point manages communicating among other nodes.

This undertaking deals with a Mobile Adhoc Network Protocol called Adhoc on demand distance vector routing ( AODV ) . AODV is a reactive protocol based on distance vector routing [ 1 ] . A radio web comprises of several radio web interfaces that are interconnected with each other on a wireless medium over which transmittals and response of information is carried out. A Wireless Network Interface is the combination of the physical constituents of a radio web arranger and the logical constituent of the Wireless Network Connection linked up with that arranger. It is the entity to which the radio connectivity and security scenes apply.

Communication is critical in sensitive scenarios. If one nexus among the pass oning nodes fails, the nodes must restart communicating utilizing other links. It merely depends on the handiness of other links ( i.e interfaces ) every bit good as the support of these multiple interfaces. If a web protocol has the support of multiple interfaces so connectivity or communicating can be guaranteed even when one peculiar nexus fails. AODV protocol works good for one physical interface. In our survey we aim to implement AODV protocol that supports more than one physical interface.

1.1 PURPOSE STATEMENT

Suppose the nodes of a web are available with multiple physical interfaces over which they can pass on individually. If we are reassigning the informations between two nodes of a web utilizing an interface and the corresponding nexus fails, there will be a data loss and we will hold to retransmit the informations utilizing other interface.

Bacillus

Link Type 1

A

Link Type 2

Figure 1.1 Concept of Multiple Interfaces between Nodes

In another instance, say two nodes of a web are connected over a peculiar interface ( nexus ) , and 2nd node is connected to a 3rd node over another type of interface. Now suppose first node has to direct informations to 3rd node via 2nd node, as shown in the figure below.

A

Bacillus

C

Link Type 1

Link Type 2

Figure 1.2 Concept of Different type of Interfaces

In this state of affairs, 2nd node must be capable of covering with multiple physical interfaces while running a individual web protocol.

1.2 Aim

To supply with guaranteed communicating for mission critical applications where connectivity among pass oning nodes is required in any instance.

To guarantee informations reassign among pass oning nodes of Mobile Ad Hoc Network even if one of physical nexus fails.

To do a web independent from type of interfaces ( links ) between the nodes of that web.

1.3 ORGANIZATION OF STUDY

This papers comprises of four chapters. Chapter one includes an debut to Manets.

Chapter two includes an overview to mantids routing protocols sorting the generic classs i-e proactive, reactive, intercrossed etc.

Chapter three includes complete account of adhoc on demand distance vector ( AODV ) routing protocol depicting the path find, package format, traverse of a package and complete methodological analysis of AODV.

Chapter four contains execution of AODV in Opnet depicting the nodes used in Manets and how to configure the properties of the specified nodes.

Chapter 2

LITERATURE CITED

2. 1 Manet

2. 1. 1 Concept

A nomadic ad-hoc web ( MANET ) is a type of radio ad-hoc web, that is a self-configurable web dwelling of nomadic routers and associated hosts that are connected by wireless links. It is an independent system in which routers are free to travel indiscriminately and form themselves randomly therefore, the web ‘s radio topology may alter quickly and erratically. Such a web may work in a standalone manner, or may be connected to a bigger web.

Wireless ad-hoc web is a web in which the communicating links are wireless. In an ad hoc web each node is willing to send on informations for other nodes, so the nodes send oning informations are determined dynamically. This is in contrast to wired web engineerings in which designated nodes, such as routers, switches, hubs, and firewalls, execute the undertaking of send oning the information. It is besides in contrast to pull off wireless webs, in which a particular node known as an entree point manages communicating between other nodes.

The traffic types in ad hoc webs are rather different from those in an substructure radio web, including:

1 ) Peer-to-Peer: In this type of substructure communicating between two nodes are within one hop. Therefore web traffic ( Bps ) is normally consistent.

2 ) Remote-to-Remote: It allows communicating between two nodes beyond a individual hop while keeping a stable path between them. It is merely possible because of several nodes remaining within communicating scope of each other in a individual country or perchance traveling as a group.

3 ) Dynamic Traffic. In this design nodes are dynamic and nomadic. Therefore paths must be reconstruct. This consequences in a hapless connectivity and web motion in short explosions.

There are cardinal differences among the architecture of wired webs and radio webs. The obvious difference is that nodes in ad hoc webs are nomadic, there are a figure of less obvious but every bit important differences. The bandwidth nowadays is of the order of 1 Mbps, an order of magnitude less than that of wired webs. Second, all communicating in a radio web is broadcast, which means that broadcast is no more dearly-won than unicast. And eventually, wireless links are much more error prone comparison to wired links. Figure shows the illustrations of both substructure and infrastructureless ad hoc radio webs.

Figure 2.1 Difference between Infrastructure based Network and Ad hoc Network

2. 1. 2 Typical APPLICATIONS OF MANET ‘S

Minimal constellation and fast deployment make ad hoc webs suited for exigency state of affairss like natural catastrophes or military struggles. The decentralised nature of most wireless ad hoc webs makes them suited for a assortment of applications where cardinal nodes can non be relied on, and may better the scalability of wireless ad-hoc webs compared to wireless managed webs. Applications for MANETs are wide ranging and have use in many critical state of affairss:

Figure 2.2 Applications in Military Communications

Rescue Operationss

An matchless application is for hunt and deliverance operations. These scenarios are characterize by the deficiency of installed communications substructure. This may be because all of the equipment was destroyed, or perchance because the part is excessively distant. Saviors must be able to pass on in order to do the most first-class usage of their energy, but besides to keep safety. By automatically making a information web with the communications equipment that the saviors are already transporting, their occupation could be made easier.

Scalability

A commercial intent for MANETs includes omnipresent calculating. By allow computing machines to send on informations for others, informations webs may be extended far beyond the usual range of installed substructure. Networks may be made more extensively available and easier to utilize.

Figure 2.3 Other Applications of MANET ‘s

Dependability

Application of MANETs is sensor webs. This engineering is a web based of a really big figure of little detectors. These can be used to place any figure of belongingss of an country. Examples include force per unit area, temperature toxins, pollutions, etc. The capableness of each detector is really limited, and each must trust on others in order to send on informations to a chief computing machine. Individual detectors are limited in their computer science capacity and are prone to failure and loss. Mobile ad-hoc detector webs could be the key to the hereafter fatherland security.

More Applications include personal country networking where the nomadic nodes may be cell phones, laptops, etc. They besides have a great potency in military operations where the nodes may be soldiers, armored combat vehicles, or aeroplanes. In add-on they could turn out to be utile in urban environments every bit diverse as cab cab webs, conference suites, boats and ships, hunt operations every bit good as policing and firefighting.

2. 2 MANET ROUTING PROTOCOLS

Mobile Ad hoc web became a hot subject for research particularly for notebook and other handheld nomadic devices like PDA ( Personal Digital Assistant ) , cell phone etc in the mid to late 90s. Up till now many research workers contributed their attempts in the MANET by suggesting their ain protocols. Hundreds of protocols have been proposed for MANET that works good under specific web substructure and demands. There is no individual protocol that would suit for all the scenarios with different web nodes, node mobility form and traffic tonss. To get the better of this job we classify the major protocols under assorted classs as listed below.

2.2.1 Proactive Routing ( Table-Driven )

This type of protocols maintains updated lists of finishs and their paths by administering routing tabular arraies throughout the web clip to clip. Some major protocols of this class are:

WRP

DSDV

2.2.2 Reactive Routing ( On-Demand )

This type of protocols finds a path on demand by deluging the web with Route Request packages. This protocol need non to keep the paths so cut down routing operating expense is observed. Three different sorts of messages are go arounding among the web [ 2 ] i.e. path petition ( RREQ ) to bespeak a new path when one node want to direct informations, path answer ( RREP ) after the successful find of finish node so the path answer message is frontward back to the beginning node and path mistake ( RERR ) is use when an mistake occurred in the web like nexus broken etc. Some major protocols of this class are:

DSR

AODV

2.2.3 Hybrid Routing ( blend of Reactive and Proactive )

Hybrid Routing, normally referred to as balanced-hybrid routing, is a combination of distance-vector routing, which works by sharing its cognition of the full web with its neighbours and link-state routing which works by holding the routers tell every router on the web about its closest neighbours. Some major protocols of this class are:

SSR

ZRP

2.2.4 Hierarchical ( Zone/Cluster-Based ) Routing Protocols

This type of protocols frequently group routers together by map into a hierarchy [ 3 ] . Its Algorithm is based on nexus province [ 4 ] . The routing is ab initio established with some proactively prospected paths and so serves the demand from to boot activated nodes through reactive implosion therapy on the lower degrees. Some major protocols of this class are:

DDR

Galvanic skin response

2.2.5 Geographic Routing Protocols ( Location based )

Geographic routing is besides known as georouting or positionbased routing. Geographic Routing Protocols relies on geographic place information. The chief thought is to direct the message from beginning to finish based on geographic location alternatively of web reference. The beginning node is cognizant of the location of the finish. So there is no demand to detect path and any extra cognition of web topology. Geographic routing is based on two chief premises:

It is assumed that every node knows it ain and its web neighbours places.

The beginning of a message is assumed to be informed about the place of the finish.

Georouting is really singular because it operates without any routing tabular arraies. Furthermore, one time the location of the finish is know, all operations are strictly local, that is, every node is required to maintain path merely of its direct neighbour.

Geographic routing optimizes route computation mechanism by connoting the significance of spacial location with regard to web public presentation. Normally they use Global Position System ( GPS ) for routing.

2.2.6 Power-Aware Routing Protocols

Ad hoc nodes are operated by battery and have limited energy assets, because a node in an ad hoc web is usually a laptop, a personal digital helper or any nomadic device, energy supplied by batteries is likely to be a deficient resource, and in some applications energy is wholly non-renewable [ 5 ] , do energy efficiency a cardinal concern in the operation of such webs specially during wireless connectivity. Furthermore the life-time of batteries has non been improved as fast as treating velocity of microprocessors. In MANET the node consume more energy as comparison to other radio web because of absence of web substructure. Mobile node in MANET must move as a router and articulation in the procedure of packages send oning so packets tonss are high. Now we will discourse power direction schemes Figure.

Fig 2.4 Provides an overview of the power cognizant direction strategy.

Battery Management: The challenge is non to supply each node with higher battery power but to use the available battery power in a really efficient mode.

Transmission Power Management: Find out the minimal power restriction path. Find out the maximal mean power path.

System Power Management: Find out the ways to minimise the power ingestion during the processing undertaking.

2.2.7 Multicast Routing Protocols

Multicast routing protocols allow a aggregation of multicast routers to build distribution trees when a host on a straight attached subnet, normally a LAN, wants to acquire traffic from a certain multicast group. Some major protocols of this class are:

DVMRP

MOSPF

2.2.8 Adaptive Routing ( Situation Aware )

This type of protocols combines the advantages of proactive and of reactive routing. The routing is ab initio established with some proactively prospected paths and so serves the demand from to boot activated nodes through reactive implosion therapy. [ 6 ]

2.3 LOCAL AREA NETWORK ( LAN )

Local country webs ( LANs ) are computing machine webs runing in size from a few computing machines in a individual office to 100s or even 1000s of devices spread across several edifices. They function to associate computing machines together and supply shared entree to pressmans, file waiters, and other services. LANs in bend may be plugged into larger webs, such as larger LANs or broad country webs ( WANs ) , linking many computing machines within an organisation to each other and/or to the Internet. [ 113 ]

You do n’t needfully hold merely computing machines on a LAN. You can besides link pressmans, difficult discs, CD-ROMs, Printers, Modems, etc and other devices for usage by other computing machines on the web as if they were their ain. For case, if you connect a pressman on a LAN and configure it to be shared among all users on the LAN, print occupations can be sent to that pressman from all computing machines on the LAN. Computers that offer resources are called Waiters.

Computers called Workstations can attach the resources ( typically hard discs and pressmans ) offered by waiters as if they were their ain.

A computing machine can be both a Server and Workstations at the same clip, in which instance it is called a Peer. Networks without dedicated waiters are called peer-to-peer webs. Networks with one or more dedicated waiters are called server based webs even though they may besides hold equals on them.

Because the engineerings used to construct LANs are highly diverse, it is impossible to depict them except in the most general manner. Cosmopolitan constituents consist of the physical media that connect devices, interfaces on the single devices that connect to the media, protocols that transmit informations across the web, and package that negotiates, interprets, and administers the web and its services. Many LANs besides include signal repeaters and Bridgess or routers, particularly if they are big or connect to other webs.

The undermentioned features differentiate one LAN from another:

Topology: The geometric agreement of devices on the web. For illustration, devices can be arranged in a star, pealing or in a consecutive line.

Figure 2.5 Star Network Topology and Bus Topology

Protocols: The regulations and encoding specifications for directing informations. The protocols besides determine whether the web uses a peer-to-peer or client/server architecture.

Figure 2 Peer to Peer Architecture and Clint Server Architecture

Figure 2.6 Peer to Peer and Client Server Architecture

Media: Devicess can be connected by twisted-pair wire, coaxal overseas telegrams, or fiber ocular overseas telegrams. Some webs do without linking media wholly, pass oning alternatively via wireless moving ridges.

LANs are capable of conveying informations at really fast rates, much faster than informations can be transmitted over a telephone line ; but the distances are limited, and there is besides a bound on the figure of computing machines that can be attached to a individual LAN.

2.4 WIRELESS LOCAL AREA NETWORK

A wireless local country web ( WLAN ) is a local country web ( LAN ) that does n’t trust on wired Ethernet connexions. A Wireless local area network can be either an extension to a current wired web or an alternate to it. Use of a WLAN adds flexibleness to networking. A WLAN allows users to travel about while maintaining their computing machines connected.

Figure 2.7 Wireless Local Area Network

WLANs have informations transportation velocities runing from 1 to 54Mbps, with some makers offering proprietary 108Mbps solutions. The 802.11n criterion can make 300 to 600Mbps.

Because the wireless signal is broadcast so everybody nearby can portion it, several security safeguards are necessary to guarantee merely authorised users can entree your WLAN.

A WLAN signal can be broadcast to cover an country runing in size from a little office to a big campus. Most normally, a WLAN entree point provides entree within a radius of 65 to 300 pess.

IEEE 802.11 is a set of criterions for wireless local country web ( WLAN ) computing machine communicating, developed by the IEEE LAN/MAN Standards Committee ( IEEE 802 ) in the 5 GHz and 2.4 GHz public spectrum sets.

Figure 2.8 Wireless Local Area Network Ad Hoc Mode ( No Access point )

2.4. 1 802.11 Standards

The 802.11 household includes several hardware criterions. The most popular are those defined by the 802.11b and 802.11g protocols, and are amendments to the original criterion. 802.11a was the first radio networking standard, but 802.11b was the first widely accepted one, followed by 802.11g and 802.11n. The 802.11a, B, and g criterions are the most common for place radio entree points and big concern radio systems.

Security was originally purposefully weak due to export demands of some authoritiess. The security criterion WEP ( Wired Equivalent Privacy ) was introduced. It encrypts informations traffic between the wireless entree point and the client computing machine, but does n’t really procure either stop of the transmittal. Besides, WEP ‘s encoding degree was comparatively weak ( merely 40 to 128 spots ) . After WEP another security strategy was introduced known as WPA ( Wi-Fi Protected Access ) . This strategy implemented higher security and addresses the defects in WEP, but is intended to be merely an intermediate step until farther 802.11i security steps were developed. It was subsequently enhanced via the 802.11i amendment after governmental and legislative alterations. 802.11n is a new multi-streaming transition technique that is still under bill of exchange development, but merchandises based on its proprietary pre-draft versions are being sold. Other criterions in the household ( c-f, H, J ) are service amendments and extensions or corrections to old specifications.

802.11b and 802.11g use the 2.4 GHz ISM set. Because of this pick of frequence set, 802.11b and g equipment may on occasion endure intervention from microwave ovens and cordless telephones. Bluetooth devices, while operating in the same set, in theory do non interfere with 802.11b/g because they use a frequence skiping dispersed spectrum signaling method ( FHSS ) while 802.11b/g uses a direct sequence spread spectrum signaling method ( DSSS ) . 802.11a uses the 5 GHz U-NII set, which offers 8 non-overlapping channels instead than the 3 offered in the 2.4GHz ISM frequence set.

2.4.2 WLAN types

The private place or little concern WLAN

Normally, a place or concern WLAN employs one or two entree points to air a signal around a 100- to 200-foot radius. With few exclusions, hardware in this class subscribes to the 802.11a, B, or g criterions ( besides known as Wi-Fi ) . Home and office WLANs adhering to the new 802.11n criterion are looking.

The endeavor category WLAN

This type employs a big figure of single entree points to air the signal to a broad country. The entree points have more characteristics than equipment for place or little office WLANs, such as better security, hallmark, distant direction, and tools to assist incorporate with bing webs. These entree points have a larger coverage country than place or little office equipment, and are designed to work together to cover a much larger country. Such equipment adheres to the 802.11a, B, g, or n standard.

Wireless WAN ( broad country web )

Although a WAN by definition is the exact antonym of a LAN, radio WANs ( WWANs ) deserve brief reference here, particularly because the differentiation is going less and less obvious to stop users. WANs used to be in order to link LANs in different geographical countries.

Chapter 3

3 ADHOC ON DEMAND DISTANCE VECTOR ( AODV ) ROUTING PROTOCOL

AODV ( Ad-hoc On-demand Distance Vector Routing ) is a loop-free routing protocol for Ad hoc webs. It is designed to be self-starting in an environment of nomadic nodes, withstanding a assortment of web behaviors such as node mobility, nexus failures and package losingss. It is a reactive routing protocol, intending that it establishes a path to a finish merely when required. The most common routing protocols of the Internet are proactive, intending they find routing waies independently of the use of the waies.

AODV maintains a routing tabular array at each node. The routing table entry for a finish contains following indispensable Fieldss:

• Destination IP Address

• Prefix Size

• Destination Sequence Number

• Next Hop IP Address

• Lifetime ( termination or omission clip of the path )

• Hop Count ( figure of hops to make the finish )

• Network Interface

• State and routing flags ( e.g. valid, invalid )

All packages designated to the finish are sent to the following hop node. The sequence figure acts as a signifier of time-stamping, and is a step of the freshness of a path. The hop count interprets the current distance to the finish node.

3.1 ROUTE DISCOVERY IN AODV

In AODV, nodes detect paths in request-response rhythms. A node requests a path to a finish by airing an RREQ ( Route Request ) message to all its neighbours. When a node receives an RREQ message but does non hold a path to the called for finish, it in bend broadcasts the RREQ message. Besides, it retrieves a reverse-route to the requesting node which can be used to send on subsequent reactions to this RREQ.

This procedure repeats until the RREQ reaches a node that has a valid path to the finish. This node ( which can be the finish itself ) responds with an RREP ( Route Reply ) message. This RREP is unicast along the contrary paths of the intermediate nodes until it reaches the requesting node. Therefore, at the terminal of this request-response rhythm a bidirectional path is accomplished between the bespeaking node and the finish.

When a node loses connectivity to its following hop, the node invalidates its path by directing an RERR ( Route Error ) to all nodes that have potentially received its RREP.

On reception of the AODV messages: RREQ, RREP and RERR, the nodes update the following hop, sequence figure and the hop counts of their paths.

3.1.1 Flow of Events in AODV

Following is a flow chart which shows the activities of nodes in the web and the sequence of those activities when they receive AODV messages.

Figure 3.1 Flow Chart depicting responses of Nodes on having different AODV messages

3.1.2 Example

An illustration is presented over here to understand the path find in AODV.Suppose S would wish to pass on with D.

Figure 3.2 AODV Route Development measure 1

The node broadcasts a RREQ to happen a path to the finish. S generates a Route Request with finish reference, Sequence figure and Broadcast ID and sent it to his neighbour nodes.

Figure 3.3 AODV Route Development measure 2

Each node having the path petition sends a path back ( Forward Path ) to the node.

Figure 3.4 AODV Route Development measure 3

A path can be determined when the RREQ reaches a node that offers handiness to the finish, e.g. , the finish itself ) .

Figure 3.5 AODV Route Development measure 4

The path is made available by unicasting a RREP back to D and is written in the routing tabular array from S. After having the path answer each node has to update its routing tabular array if the sequence figure is more recent.

Figure 3.6 AODV Route Development measure 5

Now node S can pass on with node D.

Figure 3.7 AODV Route Development measure 6

When a nexus interruption in an active path is discovered, the broken nexus is invalid and a RERR message is sent to other nodes. If the nodes have a path in their routing tabular array with this nexus, the path will be wiped off. Node S sends one time once more a path petition to his neighbour nodes, or a node on the way to the finish can seek to happen a path to D. This mechanism is called: Local Route Repair.

Figure 3.8 Coevals of Route Error on nexus interruption

3.3 ADVANTAGES AND DISADVANTAGES

The chief advantage of this protocol is that paths are established on demand and finish sequence Numberss are utilised to happen the most latest path to the finish. The connexion apparatus hold is minor. One of the disadvantages of this protocol is that intermediate nodes can ensue in inconsistent paths if the beginning sequence figure is really old and the intermediate nodes have a higher but non the latest finish sequence figure, thereby keeping stale entries. Besides multiple Route Reply packages in response to a individual Route Request package can take to heavy control overhead. Another disadvantage of AODV is that the periodic beaconing consequences in unneeded bandwidth ingestion.

3.4 RECEVING RREQ ( Route Request )

When a node receives RREQ, it checks whether it has an entry for the conceiver of RREQ in its routing tabular array or non. If an entry is non present, it saves the path to the conceiver with sequence figure present in RREQ message. If an entry is already present in its routing tabular array for the conceiver, so it compares the bing sequence figure in routing tabular array with the standard 1. If the standard sequence figure is greater than the bing one it updates the path entry for conceiver and so processes this RREQ. If the standard sequence figure is smaller or equal to the bing one it does non update the path entry for conceiver and so processes this RREQ.

The receiving node so compares its path RREQ ID with the RREQ ID ‘s of RREQ ‘s which it has already received ( seek list ) . If RREQ ID lucifers, it discards RREQ and does non send on it.

If RREQ ID does non fit with any entry of the list of RREQ ID ‘s, so Further the having node of RREQ cheques whether it is the finish of that RREQ or non. If it is the mark finish, it generates RREP ( Route Reply ) in response and unicasts it to the conceiver. if it is non the mark finish of that RREQ, it checks its routing tabular array to see whether it contains the path to the mark finish or non. Besides this path is fresh or non ( by comparing sequencing figure ) . If it finds a fresh path to finish in its routing tabular array so in response it generates RREP and unicasts it to the conceiver of RREQ. If it does non incorporate a fresh path to the mark finish so it broadcasts this RREQ on its all interfaces.

In instance of multiple physical interfaces, every node trades with RREQ received on a peculiar interface independently and individually. It means, when a node receives RREQ on interface 1, it will treat it as mentioned above but when it will have the same RREQ ( with same RREQ ID ) through other interface, the node will fling this petition after salvaging rearward path to originator through this way.

While this node will direct RREP of this RREQ it will unicast it on both interfaces.

3.5 RECEVING RREP ( Route Reply )

When a node receives a RREP message, it foremost updates its forward path to the immediate transmitter of RREP. Then it checks whether this RREP is the response of a RREQ generated by itself or non. If the standard RREP is of its ain RREQ, the procedure of path find is completed here. Now it starts directing informations to the desired finish. If this RREP is non of its RREQ so it unicasts RREP to the following hop towards finish.

In instance of multiple physical interfaces, when a node receives RREP on an interface, it updates its forward path matching to that interface and unicasts RREP to following hop on the interface over which it antecedently received corresponding RREQ.

RECEVING RERR ( Route Error )

When a node receives RERR message, it removes matching affected paths. If the receiving node is the following hop of any other node towards the finish with affected nexus interruption, so it forwards RERR to its following hops so that they can besides take their affected paths ; else it discards RERR.

In instance of multiple physical interfaces, RERR will be generated and forwarded merely when all links to a node fail because it can be a instance that one nexus to a node may neglect but other may work mulct.

3.7 Example

An illustration is presented over here to understand the node based AODV routing. This illustration covers tracking and processing of Route Request, Route Reply and Route mistake messages. In this illustration a scenario is shown in which four nodes are connected as shown:

Guess:

Suppose a web of four Nodes as shown in diagram below.

Node 1 has one interface IF-1 and its neighbour is Node 2.

Node 2 and Node 3 are neighbours on both interfaces IF-1 and IF-2.

NODE 1

IF-1

IF-1

NODE 4

IF-1

IF-1

IF-2

IF-2

NODE 2

NODE 3

IF-1

IF-1

IF-2

IF-2

IF-1

IF-1

IF-2

IF-2

Similarly Node 3 and Node 4 are neighbours on both interfaces IF-1 and IF-2.

Figure 3.9 Example Wireless Ad hoc Network of four Nodes

Sequence of Events in Route Discovery of Node 4 from Node 1:

Types of messages used to develop AODV routing:

RREQ: Path Request

RREP: Path Reply

RERR: Path Mistake

3.7.1 Traversing of RREQ:

Processing of Node 1

Node 1 will bring forth a RREQ with a sequence figure and wide dramatis personae this RREQ on its interface IF-1. Node 1 is merely listenable by Node 2 and merely on IF-1.

Processing of Node 2

Node 2 will have RREQ on its interface IF-1.

Node 2 will put change by reversal path for Node 1 on interface IF-1 if the sequence figure of standard RREQ is greater than the bing one.

Node 2 will look into if it is the mark finish or non.

If Node 2 is the finish node, it will update its sequence figure for Node 1 ( if necessary ) , generate RREP and will unicast it to Node 1 on interface 1.

If Node 2 is non the finish so it will look into whether it has any path to finish or non.

If it already has the path to the finish on its both interfaces or any interface, it will compare the bing sequence figure and that in the RREQ. If the bing sequence figure is equal or greater than the standard one, Node 2 will bring forth and unicast RREP to Node 1 on its interface 1.

If the bing sequence figure is less than the standard one, Node 2 will air this RREQ on its both interfaces IF-1 and IF-2. Before wide projecting the RREQ on either interface, Node 2 will put the right interface index in RREQ.

In this illustration, since Node 2 will non hold path to Node 4, so it will wide dramatis personae received RREQ.

Processing of Node 3

Node 3 will have RREQ on its both interfaces IF-1 and IF-2.

Suppose Node 3 will have RREQ on interface IF-1 foremost.

Node 3 will put change by reversal path for Node 1 on interface IF-1 if the sequence figure of standard RREQ is greater than the bing one.

Node 3 will look into if it is the mark finish or non.

If Node 3 is the finish node, it will update its sequence figure for Node 1 ( if necessary ) , generate RREP and will unicast it to Node 2 on interface 1.

If Node 3 is non the finish so it will look into whether it has any path to finish or non.

If it already has the path to the finish on its both interfaces or any interface, it will compare the bing sequence figure and that in the RREQ. If the bing sequence figure is equal or greater than the standard one, Node 3 will bring forth and unicast RREP to Node 2 on its interface 1.

If the bing sequence figure is less than the standard one, Node 3 will air this RREQ on its both interfaces IF-1 and IF-2. ( Does non hold fresh plenty path ) . Before wide projecting the RREQ on either interface, Node 3 will put the right interface index in RREQ.

Node 3 will besides have this RREQ on interface IF-2 once more. If Node 3 has already flooded this RREQ received on interface 1, it will fling this RREQ, but later it will besides direct RREP to Node 2 on this Interface 2. Besides Node 3 will put change by reversal path for Node 1 on interface IF-2.

In this illustration, since Node 3 will non hold path to Node 4, so it will wide dramatis personae received RREQ on its both interfaces 1 and 2.

Node 3 is merely listenable by Node 4 on its both interfaces IF-1 and IF-2.

Processing of Node 4

Same stairss will be repeated by Node 4 as by Node 3.

In this illustration Node 4 will bring forth RREP.

Traversing of RREP:

Processing of Node 4

Suppose Node 4 receives RREQ on interface 1 foremost.

First Node 4 will put change by reversal path to Node 3 on interface 1. Then it will update its ain sequence figure by comparing its sequence figure with the sequence figure of finish in standard RREQ.

Then it will bring forth and unicast RREP to Node 3 on interface 1.

When Node 4 will once more have same RREQ through other interface, it will put its contrary path to Node 3 through this interface and will fling it.

Processing of Node 3

Suppose Node 3 will have RREP on interface 1 foremost.

When Node 3 will have RREP message, it will put its forward path to Node 4 on interface 1.

Then Node 3 will look into if it is the conceiver of matching RREQ or non.

If it is the conceiver of matching RREQ, it will get down directing the information packages queued to the node from which it late received RREP.

If it is non the conceiver of matching RREQ of received RREP so it will unicast ( frontward ) this RREP to following hop through the interfaces over which it received matching RREQ.

In this illustration, Node 3 will unicast this RREP to Node 2 on interfaces 1.

Processing of Node 2

Suppose Node 2 will have RREP on interface 1 foremost.

When Node 2 will have RREP message, it will put its forward path to Node 3 on interface 1.

Then Node 2 will look into if it is the conceiver of matching RREQ or non.

If it is the conceiver of matching RREQ, it will get down directing the information packages queued to the node from which it late received RREP.

If it is non the conceiver of matching RREQ of received RREP, so it will unicast ( frontward ) this RREP to next hop ( Node 1 ) through interface 1 merely.

In this illustration Node 2 will unicast this RREP to Node 1 on interface 1.

Processing of Node 1

When Node 1 will have RREP message, it will put its forward path to Node 2 on interface 1.

Then Node 1 will look into if it is the conceiver of matching RREQ or non.

If it is the conceiver of matching RREQ, it will get down directing the information packages queued to the node from which it late received RREP.

In this illustration, since Node 1 is the conceiver of RREQ so it will now get down directing informations to Node 2 destined for Node 4.

Therefore the Route Discovery procedure completes.

In this illustration the routing tabular array of Node 1 will look like:

Finish node

Following node

Hop Count

My Interface index

Following node IP

Destination Sequence Number

1

1

1

1

IP-1

Ten

3

3

1

1

IP-1

Ten

4

3

2

1

IP-1

Ten

The routing tabular array of Node 2 will look like:

Finish node

Following node

Hop Count

My Interface index

Following node IP

Destination Sequence Number

2

2

1

1

IP-1

Ten

3

2

2

1

IP-1

Ten

4

2

3

1

IP-1

Ten

Flags, provinces and timers are non shown in the tabular array.

Data type of node and interface index is 8 spot whole number.

CHAPTER4

4.1 Execution

We have chosen Opnet environment in order to implement our undertaking. Opnet provides a complete environment for developing communicating networks.Opnet patterning undertaking falls into three stages of i-e informations aggregation, simulation and analysis. The bundle consists of a figure of tools, each one focussing on peculiar facets of the mold undertaking. These tools fall into three major classs that correspond to the three stages of mold and simulation undertakings: Specification, Data Collection and Simulation and Analysis. These stages are necessarily performed in series. They by and large form a rhythm, with a return to Specification following Analysis. Specification is in fact divided into two parts: initial specification and re-specification.

Figure 4.1 Simulation Project Cycle of OPNET

4. 1. 1 Manets Node Model

Figure 4.2 Nodes of MANET ‘s

Wireless LAN workstations and waiters

Node theoretical accounts can be used to do application traffic such as Email, VOIP, HTTP, TCP over IP and can besides be configured to run AODV as the routing protocol.

MANET Stationss

These node theoretical accounts can be used to bring forth natural packages over IP over WLAN. They can configure as a traffic beginning or finish and can be configured to run AODV as the routing protocol.

Profile Config

Profiles config describes the activities of a user or group of users in footings of the applications used over a interval of clip. Different profiles can be created that can stand for different user groups.

Application Config

A profile is created utilizing different application definitions ; for every application definition, you can stipulate managing parametric quantities such as start clip, continuance and repeatability.

Rxgroup Config

Rxgroup Config is used to cipher set of possible receiving system that a node can pass on with, this public-service corporation node can significantly rush up a simulation by extinguishing receiving systems that do non fit.

Undertaking Config

Task confrigration nodes are used to configure usage Applications

Mobility Config

These nodes are used to specify mobility profile. Motion of the object ( node ) Is controlled by this node based on configure parametric quantities.

4. 1.2 AODV CONFIGURATION in OPNET

AODV can be configured by redacting the properties of a node.

Figure 4.3 Attribute of a Node

4. 1.3 AODV web modeled in Opnet

The web with five Wireless LAN nodes is deployed. All the nodes in the web are configured to work under ad hoc manner. Among the five nodes as shown in figure four nodes are fixed ad hoc nodes while one node ( mobile_node ) is nomadic. The mobile_node starts traveling after 200 seconds along the way specified during the simulation period.

Figure 4.4 Configured five node campus web for TCP traffic

After deploying the nodes they are to be configured and the undermentioned constellation is to be done:

Undertaking Configuration, Application Configuration, Profile Configuration, Mobile node

Once you are done with the constellation consequences are analyzed by running distinct event simulation.