Comparisons Of Mobile AD HOC Routing Protocols Computer Science Essay

Published:

In today's rapid world mobility is very important, user needs reliable and fast connectivity while on the move i.e. mobile ad hoc network (MANET) which is preferred among cellular and MANET. Performance can vary in different situation. So, to avoid improper functioning of the network a good routing protocol should work equally well in small, medium and large networks on Different speed and different location as user is continually changing location. It has been 40 years since when internet was introduced. Many new ways of communication has been developed to make communication faster and reliable and wireless networks are one of these developments. Wireless network are easy to deploy and less expensive. Due to movement of host and routers the network topology can be dynamic and unpredictable. Due to the dynamic nature of the MANET traditional routing protocol cannot be used in mobile ad hoc networks [11].

The purpose of this project is getting a deeper understanding of MANET and MANET protocols and to learn the working of simulation engine to simulate three protocols. Namely, Ad Hoc On Demand Distance Vector Routing Protocol (AODV), Optimized Link State Routing Protocol (OLSR) and Dynamic Source Routing Protocol (DSR). The performance of these protocols are analyzed and compared in different real life situation. Five different statics (Data Drop because of buffer overflow, Data Drop because of retry threshold, Delay, Load and throughput) will be considered as performance criteria for analyzing the performance of the three protocols in small, medium and large MANET.

RECIPIENT FOR RESEARCH

Lady using a tablet
Lady using a tablet

Professional

Essay Writers

Lady Using Tablet

Get your grade
or your money back

using our Essay Writing Service!

Essay Writing Service

The recipient for my research will be:

My self

My research adviser

My institution, Dublin Business School

The management and staff in Response Group XYZ LTD.

SUITABILITY OF THE RESEARCHER TO THE RESEARCH

I am an computer engineer having a strong background on the development of networking and hardware as I had worked as network administer in NIIT Hyderabad and had over one year of experience. Presently I am pursuing my MSC in computing and information system which has built the basic platform for me to do the positive research. I can collect the data and develop theories from data and subsequently relate them to the literature

LITERATURE REVIEW

WIRELESS MOBILE AD HOC NETWORKS

INTRODUCTION OF MANET

A mobile ad hoc network is a self configured network of mobile devices connected by wireless links. It is also called mobile mesh network. Each device is free to move anywhere and in any direction in MANET. Devices in MANET change their links with other devices very frequently and these devices forward the traffic which is not for them. The most difficult job in MANET is to make devices capable of routing traffic independently. These devices can communicate with each other or can connect to internet as well. [1] In ad hoc networks each device has to not only discover the neighboring device but also extract all it attributes and for this purpose it has to exchange data packets to get information and this depends on the battery life of nodes in ad hoc networks as nodes vary in MANET. So power consumption is a major challenge in mobile ad hoc networks. Most current research in the field wireless networking made it possible for mobile nodes to function in the presence of network which has infrastructure where these devices will be able to exchange the data as these devices use to exchange the data before mobile-hop cellular network (MCNS) [5]. A good example of such sort of networks is Self-Organizing Packet Radio Ad hoc Networks with Overlay (SOPRANO) [1].

As it is clear that much effort has been put in the field of ad hoc networks but still a huge research work is going on in the field of MANET .Current research is going on in field of providing standard for real time application and much research is going on quality of service field in ad hoc networks.

TYPES OF MOBILE NETWORKS

There are two types of mobile networks, mobile ad hoc networks and mobile cellular networks. In cellular networks each device is connected and controlled by a base station and shares information from other neighboring base stations. But, in case of ad hoc networks each device is independent and can configured and organize themselves. In cellular networks base station is responsible to route the packets across the network where as in ad hoc networks each is responsible for routing the traffic across the network. Mobile ad hoc network is more complex as the network can function as router as well as host. . Ad hoc networks face hostile environment like more noise and connectivity problems. Ad hoc networks are cost effective and they need less time to set up while cellular network are expensive, need more planning and they have more stable connectivity [1].

Lady using a tablet
Lady using a tablet

Comprehensive

Writing Services

Lady Using Tablet

Plagiarism-free
Always on Time

Marked to Standard

Order Now

C:\Documents and Settings\M A KHAN\My Documents\My Pictures\adhocnet.gif C:\Documents and Settings\M A KHAN\My Documents\My Pictures\cellular1.gif

Mobile Ad hoc network [7] Mobile Cellular network [8]

CHALLENGES IN AD HOC NETWORKS

There are many challenges in mobile ad hoc networks. Some of these challenges are as follows:

Scalability

Limited number of nodes is not a problem in ad hoc networks but as soon as number of nodes increase in ad hoc networks, they affect the throughput. Many nodes or nodes over the limit are a difficult challenge for mobile ad hoc networks as there is no central controller in MANET.

Security & Privacy

Protecting network is a major challenge in mobile ad hoc network because they donor have a central manager so each device is responsible for its security. There are additional steps required to allow make only authenticated devices enter in the network. MANET is an easy target for different types of attacks.

3. multicasting:

In the network, data is carried using data packets and are having different types of packets to carry different types of data. So, multicast routing is required to transfer the data.

COMMON ISSUES AND CHALLENGES FOR MANET PROTOCOLS

Handling dynamic connectivity and making sure that packets are delivered are core things for routing protocols in wireless networks. Mobile ad hoc networks have the issue of mobility as the device keep on changing their location and there is no central point which can monitor their movements and keep the routing table up to date so each other device each device has the issue of dealing with the problem of changing location of each other device. Network topology change is quite frequent in MANET so routing protocol should be capable enough to update the routing table as soon as a change happens. There is a requirement of a protocol that can response to changes in network topology quickly to make routing efficient in mobile ad hoc networks. In ad hoc networks devices moves from one position to another position and because of their movement resources are not always available. MANET has very limited bandwidth availability and a considerable problem for routing protocol.

REQUIREMENTS FOR A PERFECT ROUTING PROTOCOL

There are some requirements which are required for a good routing protocol. Most of them are based on the challenges described above. Protocols used in the wired networks cannot full fill the requirements stated above because these protocols are not designed to handles above stated issues as issues do not exist in wired networks. An ideal protocol for mobile ad hoc networks should have the following requirements.

Resource usage by routing protocol should be minimum for performing the task of routing data. Resources include battery life bandwidth and others. [12]

Time sensitive application should be provided with the good quality of service as in these applications a small delay can be very critical and can be damaging. [12]

Routing protocol is required to quickly exchange and update the routing table when the network topology changes and it should be as soon as a change happens. [12]

To share and update the routing information, the number of devices should be minimum as this way less bandwidth will be consumed and less time will be required for routing updates and changes.[12]

Protocol should not work in a centralized fashion as if it will be centralized then more overhead will be there when there will be a requirements of attaining a network topology change and it can suffer from the problem of single point failure. Protocol should work in distributed fashion as in this approach the problem of single point failure is not there and if one device is not working then other devices can cover that device work making system a fault tolerant system. If system will be fault tolerant then it can be deployed in many critical applications.[12]

MOBILE AD HOC ROUTING PROTOCOLS CLASSIFICATION

Routing protocol for ad hoc networks are classified in different classses based on different criteria. Commonly used classes use information which is based on routing topology, usage of specific resources, temporary information which is available for routing and update mechanisam. Use of this information is not an hard and fast rule some time some of the protocol can use information abailaable in criteria and some time some of them can use combination of two criteria based on the approch which is followed in the development of the protocol.

ROUTING INFORMATION UPDATE MECHANISM BASED

Lady using a tablet
Lady using a tablet

This Essay is

a Student's Work

Lady Using Tablet

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

Examples of our work

A successful and efficent routinf in an ad hoc environment can be accomplished by using updated and reliable information. Route discovery and route maintaince depends on this information. One of the desirable feature in the design of the protocol is fast and efficent updates and these updates should not overload the network as in ad hoc networks limited bandwidth is available. There are three main categories so far routinf information update mechanism is concerned.

Table Driven(Pro Active ) Protocol.

On Demand (reactive) Protocols.

Hybrid protocols.

TEMPORAL INFORMATION USAGE ROUTING PROTOCOLS.

Temporal informatio is used to route the data in these protocols. Temporal information includes life time of the a selected path and link e.t.c. There are two catogory of these protocols . First one is the one which use the past temporal information and second one uses future temporal information.There are different matrics availble in network history and they can use some of them like remaining battery power and predication that a selected link will be available in future.

ROUTING TOPOLOGY BASED PROTOCOLS

These protocols can be divided into categories based on their working

Flat topology routing protocols

Grouping of nodes is not used in flat topology protocols and these proitocols are good for the smaller networks. If a node is closed enough approximately then a connection is established. Connectivity between two node is the main factor for routing between two nodes. Ecah node has equal responsability. These protocols give lower processing power and bettr wirelss spevctrum usage. But these protocols produce more overhead if network is big or if mobility is high.[14]

Hierarchical topology routing protocol

Hierarchical routing protocol consider one or more level of topology unlike the flat topology based protocols. Based on some criteria nodes are divided into clusters. Criteria can be location or function performed by the nodes. These cluser can overlap each other or they can be completely disjoint from each other.s When the network initilized then all the nodes are divided into clusters. Selection and revocation of the cluster leader algorithm startssoon after the network is divided into clusters. Different methods can be used to select a clusetr laeder like lowesr I.D.The clusetr leader is the incharge of the clusetr centrally.[14]

LIMITED RESOURCE UTILIZATION BASED ROUTING PROTOCOLS.

Ad hoc network have the condition of variable link states and different uncertain conditions.These protocl consider the limited resource factor to route the information.There are diffreent routing tecniques like power awre routing,security aware routing,geographical approches and routing by the use of efficent flooding.[14]

AD HOC ON-DEMAND DISTANCE VECTOR ROUTING PROTOCOL

This protocol is a good candidate for MANET. Because of its reactive nature this protocol can avoid use of unused routes which result in introduction of higher delay as compare to proactive protocols when it discovers a route (Perkins et al... 2003).

AODV first version was developed by the IEFT community in 2001. The base of ADODV is distance vector protocol. Every node is aware of the neighbors and the cast of reaching them in distance vector. Distance is infinite for a node which is not reachable. The whole routing table is exchanged with the neighbors periodically. AODV has demand routing nature and it has small delay. To reduce the amount of traffic on the network routes are only established when they required. Uncast, broadcast and multicast are supported by the AODV. Sequence number and registration of the cast solves the problem of loop and count to infinity. The cast is 1 for every host which is constant. To accommodate the moment of the node the routes age very fast. [16].

DYNAMIC SOURCE ROUTING PROTOCOL (DSR)

Dynamic source routing protocol was designed to use in the field of multichip wireless ad hoc networks. It is a quite simple and very efficient protocol. Network can organize and configure itself by the use of DSR. Route is discovered dynamically to deliver a packet from source to destination. [17]

The DSR is one of the on demand protocols and it finds a route to deliver a packet only if there is a request of finding a route. [12]

Each packet has the complete route in its header that contains the detail of nodes through which the data has to go through and it makes the routing of the packet loop free as well as there is no need of the up to date information for routing. The routing information in header of the packets can be cached so that if in future packet can be delivered to same destination using the cached path. [17]

There are two phases in DSR. One is how to discover a route and other is how to maintain that route.

TEMPORARILY-ORDERED ROUTING ALGORITHM (TORA)

TORA is a distributed routing algorithm. Performance of TORA is very good in large networks and the networks with limited bandwidth and dynamic resources. Topology change reaction is minimized by TORA. A very small set of nodes which are near the change are send control messages. In order to achieve this, node needs to achieve the routing information of their adjacent nodes. A directed acyclic graph which is rooted towards the destination is created by the use of the metric "height" during the process of route creation and maintenance. Height or reference level of each node is used to assign a logical direction to a link in order to achieve the DAG. For example if there is a link (i,j) then I will be upstream and j will be downstream node. Each link has one upstream or one downstream assignment which is according to the relative heights of the stations involved. There is only one sink in DAG and rest of the nodes should have at least one link which is going outwards. [18]

OPTIMIZED LINK STATE ROUTING (OLSR)

OLSR is a proactive routing protocol and it is one of the very famous protocols. INIRIA has developed this protocol and it was standardized by IETF. Shortest path algorithm is used in OLSR which is based on the hop count of metrics. The main concept in OLSR is the mechanism of broadcasting the required link information throughout the whole network. Multipoint relays are selected by a node amongst its neighbor. This is done in a way that every node which has two neighbors will receive the broadcast message. If MPRS will forward the message then it will reduce the requirement for the broadcast message. OLSR uses the optimized forwarding method for all types of broadcasts in entire network. The amount of link state information can be minimized by the use of OLSR and this is possible because only necessary link state information is required to be broadcasted to selector MPRS. [21]

OPNET MODELER SIMULATOR

Opnet is used to simulate the network to analyze the performance of the MANET. There are number of reasons for which Opnet has been selected .Opnet provides GUI and it has all the required functions described above excluding that it is not free to us. As I have the resources to get the Opnet so there is no reason for not choosing the Opnet. It has a hierarchical structure which is very good to debug the design and simulation, find errors and explaining the design is very simple. Most of the functions are designed on the diagrams and tables for the parameters .End user has to draw the diagram of the design and set the values in the parameters in the tables and results can be extracted in the form of graphs and can be compared with each others. Complex design is also possible in the Opnet. It is possible to designed entire new protocol and it provides the ability of doing process level modeling and node level modeling. Protocols which are simulated are DSR, Ad Hoc On demand Distance vector routing protocol and OLSR.

PERFORMANCE CRITERIA

Five different values are used as criteria to analyze the performance of the MANET.

Data drop (buffer overflow)

It is the size of the upper layer packets dropped by the MAC. There can be two reasons for this to happen.

If the data buffer for higher layer is full

If the size of the higher layer data packet is more than the one which is allowed.

2. Data drop (retry threshold)

Size of the data dropped by the all MAC from higher layer is because of consistently failing retransmission. This statics show the size of higher layer data which are dropped by the MAC of all nodes because of MAC has not received the ACK for those packets or fragments of those packets and retry count has reached the maximum limit.

Delay

This statics represents the delay for all the packets received by the MAC of all the wireless nodes. This delay is combination of the delay caused by the medium access delay or delay cause by medium station to forward the packet or the delay by the receiver.

load

Load is the size of the total amount of the load in bits/sec coming from upper layers to wireless layer to transfer. Load can be different for different networks.

Throughput

Throughput is the actual data which is transferred between the MAC of the two wireless nodes. Throughput is the actual bandwidth available. It is not possible to get full bandwidth as there can be many reasons for low availability of the bandwidth.

NETWORK DESIGN

Design of networks consists of three different networks i.e. small MANET, medium MANET and large MANET. These three different models are used in the analysis of the routing protocols and according to set criteria I will find out the best protocol from the above mentioned three protocols. Theses protocol have the same wireless stations but different number of nodes as MANET increases in size it introduces different problems so size can be used to create different scenarios. Appendix A will show the way a MNET station can be designed.

Small MANET

Small MANET consists of 25 ad hoc wireless nodes and these nodes are moving to different position according to time and these nodes also exchange the packets and route the packets.

Small MANET

Medium MANET

This network consists of the 50 mobile nodes. These nodes are mobile and constantly exchanging the packets. These nodes are performing the routing function as well.

Medium MANET

Large MANET

This network is considerably bigger than other two designs. The number of nodes in this is 100. All of these wireless ad hoc stations are changing the location and exchanging the packets while each station is acting as router as well.

Large MANET

SOFTWARE AND HARDWARE REQUIREMENTS

OPNET modeler 14 or latest

Internet

XP version of windows or latest

One PC with built in network card

RESEARCH METHODOLOGY

Research can be defined as the Systematic approach to solve a particular problem. According to Hancock (2002) methodology consider the information required, the research design, the method of collecting data and the source of information collected. Considering all these facts it is very important inspect the methodologies and decide to select the apt one for the dissertation

RESEARCH AREA

Increase in the usage of wireless network is increasing worldwide. But there are few errors in selecting the procedures or protocols for proper functioning of the network i.e. mobile ad hoc network (MANET) in this case. The Research area is to analyze the performance of three main routing protocols ( (Ad Hoc On Demand Distance Vector Routing Protocol (AODV), Optimized Link State Routing Protocol (OLSR) and Dynamic Source Routing Protocol (DSR)) depending on the variation on five static performance criteria in small, medium and large network. This three routing protocols are more preferred among all the protocols because the MANET has highly dynamic characteristics.

RESEARCH QUESTION:

The question I have selected for my dissertation is

"performance, analyzation and comparison of mobile ad hoc routing protocols in different types of wireless networks".

RESEARCH OBJECTIVE

(1) To study the Ad hoc networks, how they work, who is reasonable for their development and key concepts. To find out what are the practical application of the ad hoc networks and to find out what is the future of ad hoc networks.

(2) To understand how routing is performed in ad hoc networks.

(3) To understand how routing is different in ad hoc networks from normal routing protocols and what are the challenges of ad hoc routing.

(4) To understand working of different routing protocol. Studying each Protocol and analyze how these protocols discover route and maintain route. To find out what are the features and drawbacks of each protocol.

(5) To find out why the network designs are to be simulated before deploying them to real world.

(6)To learn OPNET and use it to simulate different routing protocol like OLSR, DSR and AODV.

(7)To compare the performance of different protocols like OLSR based on the simulation results.

(8)To draw a conclusion based on analysis about routing protocol and to compare the performance based on the statics from the simulation process.

RESEARCH HYPOTHESIS:

There is a possibility of obtaining same results for all the three routing protocols for MANET after performing analyses in five performing criteria or overall result will be two or more routing protocols.

The overall analysis and comparison of the protocols may lead to discovery of hidden characteristics of MANET which will intend to help design better routing protocols in future.

RESEARCH ONION

figure61

RESEARCH PHILOSOPHY

The research philosophy adopted for this dissertation will be a critical realism, as all the information and data is based on the information provided by the employees and the management in the organization. The research will be carried out depending on experimental data which will be collected from the organization. The experiment data is collected from real time functioning scenarios. Then the data is analyzed and efficient protocol is selected whose overall performance is best among the three routing protocols. The research philosophy can always influence the research development.

According to Saunders et al [13] research philosophy relates to development and nature of the knowledge. Research philosophy contains important assumptions about the way in which research is carried out.

RESEARCH APPROACH

The research approach in the study of routing protocols is inductive. As the research is carried out on the bases of deductive data provided by "ad hoc networks research and application group-IIMC (India)". Through comparing and analyzing the results in relation to the theory which is obtained from the secondary data and research will go hand in hand with primary data. The basic approach is to get the primary data from the organization and will be checked further using the secondary data. Conclusion is derived depending on both primary and secondary data and further suggestions will be made if any required.

RESEARCH STRATEGY

A research strategy is simply a general plan of how the researcher will go about answering the research questions and research objectives (Saunders et al, 2003: 288)

The research will use different strategies to collect and implementation of primary and secondary data. Research is based on experimental analysis of mobile ad hoc routing protocols. Identification of best protocol suitable to fit all size networks is to be found. So to reach the ultimate target experimental data will be acquired from "ad hoc networks research and application group-IIMC (India)" and is compared to the secondary data and to themselves. As the research is cross-sectional, the data from the survey is compared to case studies and real-time scenarios.

DATA COLLECTION, EDITING AND CODING

Primary data and secondary data:

Primary data will be collected from the survey questions from "ad hoc networks research and application group-IIMC (India)" and the data provided by the company in the form of graphs, calculations and comparison tables are collected. And interaction with the team members of the group additional data is collected in the form of interviews and questioners. The research can't be considered as qualitative or quantitative. So, the research is categorized as mixed research due to usage of both qualitative and quantitative research. Secondary data is collected from relevant sources such as books, journals, articles and electronic data base (internet).

POPULATION AND SAMPLE:

Research population is defined as the set of cases or group members being subjected to

Research (13). The research population identified for the proposed

Research includes management team, staff members, scientists, program developers associated with the group.

PROJECT MANAGEMENT COST AND TIME

The time for this plays an important role more over I had a less time for this research, but I will complete the project at any cost. I will try my level best to complete my research according to the planned time. The estimated cost for the successful completion of this dissertation work will consists of postal, printing and binding charges. Most of the work will be carried out through internet and computer.

NEW AND RELEVENT RESEARCH:

In today's world requirement of mobile is increasing rapidly so new technologies are been developed throughout. Demand for the wireless network increased because of it functionality. Mobile ad hoc networks are developed to make wireless network. But there is a gap between the network and the protocols selection because of the MANET dynamic nature. Researchers throughout the world are developing protocols for MANET.