Enhanced Hop Based Blocking Expanding Ring Search Algorithm

A thesis submitted in partial fulfillment of the requirements for the award of the degree of

Master of Technology
in

Computer Science and Engineering

By Neeshu Kumari Roll No: 118150015

Department of Computer Engineering & Applications

Institute of Engineering & Technology

GLA University Mathura- 281406, INDIA August, 2013

Department of computer Engineering and Applications GLA University, Mathura

17 km. Stone NH#2, Mathura-Delhi Road, P.O. Chaumuha, Mathura 281406 U.P (India)

Declaration
I hereby declare that the work which is being presented in the M.Tech. thesis A Reactive Route Discovery Approach In Mobile Ad Hoc Networks with Minimum Time Delay, in partial fulfillment of the requirements for the award of the Master of Technology in Computer Science and Engineering and submitted to the Department of Computer Engineering and Applications of GLA University, Mathura, is an authentic record of my own work carried under the supervision of Anant Ram, Associate Professor, GLA University, Mathura. The contents of this thesis, in full or in parts, have not been submitted to any other Institute or University for the award of any degree and are free from plagiarism.

Signature of Candidate: Name of Candidate: NEESHU KUMARI Roll. No. 118150015

Certificate
This is to certify that the above statement made by the candidate is correct to the best of my knowledge and belief.

Signature of Supervisor(s): Date: Name & Designation of Supervisor(s): Anant Ram, Associate Professor, Dept. of CEA GLA University, Mathura

ii

Dedicated To My Parents and Teachers

iii

Abstract
In mobile ad hoc networks routing protocols mainly focuses on to reduce the time delay and save energy during the discovery of the route from source to destination. In mobile ad Hoc networks, nodes communicate to each other by sending the RREQ packet. So the expanding ring search algorithm is used to synchronize the RREQ packets which help to reduce time delay and energy consumption of the entire network. Energy consumption can be reduced by controlling the flooding of packet across the network. Time delay can be reduced by sending the entire packet in order to find the destination with minimum time. There are various methodologies which work differently to minimize the time delay or to minimize the energy consumption. But most of them fail to handle both at the same time. In this thesis we proposed a new approach in which all the nodes contain the information up to two hops. So by using these informations it can find the destination as soon as possible. This strategy reduces the time delay and also energy consumption. Outcome of the proposed methodology is better than existing approaches.

iv

Acknowledgments
I would like to express my gratitude to all of those who provided me with the resources and guidance to complete my thesis. First, I am highly indebted to my supervisor Mr. Anant Ram for his guidance and constant supervision as well as for providing necessary information regarding the thesis& also for his support in completing the thesis. I sincerely thank to the Chancellor and Vice-chancellor of GLA University to provide a research oriented infrastructure. I would like to express my special gratitude and thanks to the Head of Department, Computer Engineering and Applications, Prof. Krishna Kant for giving me attention and time as and when needed. I would like to thank the faculty members of the Department, namely Prof. Charul Bhatnagar, Dr. Anand Singh Jalal, Dr. Manas Kumar Mishra, Mr. Rakesh Galav, Mr. Nitin Tyagi, Mr. Diwakar Bhardwaj and Mr. Ashish Sharma for their valuable suggestions. I am failing if I miss my special thanks to my colleagues Tanya Sharma, Surabhi Solanki and Kirti Agarwal who supported me at every step while facing several kinds of difficulties.My thanks and appreciations also go to my colleague in developing the thesis and people who have willingly helped me out with their abilities. I would like to express my gratitude towards my parents and family members for their kind co-operation and encouragement which help me in completion of this thesis.

NEESHU KUMARI GLA University, Mathura v

Contents
Certificate Abstract Acknowledgments List of Tables List of Figures 1. Introduction 1.1 1.2 1.2 1.4 1.5 1.6 Overview..................... Motivation ...................... Issues and Challenges ..... Objectives ........... Contribution of the Thesis .............. Outline of the Thesis .. ii iv v viii ix 1 1 3 3 7 7 8 9 10 9 24 24 31 31 28 32

2. Literature Review 2.1 Introduction . 2.2 Expanding ring search Approaches . 2.3 Comparative Analysis.. 2.4 Summary............... 3. Proposed Methodology 3.1 3.2 3.3 Introduction .... The Proposed Framework.... Summary......

Contents

4. Result and Discussion 4.1 4.2 Introduction ....

33 33

Results Analysis of ERS...... 33 4.3.1 4.3.2 Quantitative Result Analysis... Qualitative Result Analysis 34 38

4.3

Summary.. 40 41

5. Conclusions and Future Directions 5.1 5.2

Summary and Contributions ........... 41 Future Directions..... 41 44

References

vii

List of Tables

2.1 3.1 3.2 3.3 3.4 3.5 4.1 4.2 4.3 4.4 4.5 4.6 4.7 4.8 4.9 4.10 4.11 4.12

Comparative analysis of different approaches . Information of RREQ packet ... Algorithm for source node ... Algorithm for intermediate node . Algorithm for destination node Procedure of blocking .. Packet delivery ratio at 100 pause time ... Packet delivery ratio at 200 pause time ... Packet delivery ratio at 300 pause time ... Packet delivery ratio at 400 pause time ... Packet delivery ratio at 500 pause time ... End to end delay at 100 pause time . End to end delay at 200 pause time . End to end delay at 400 pause time . End to end delay at 500 pause time . Average End-To-End Delay Average packet delivery ratio .. Average normalized routing load viii

24 27 29 30 31 31 34 35 35 35 36 36 36 37 37 38 38 38

4.13

Simulation parameters .

40

ix

List of Figures

1.1 1.2 1.3 2.1 2.2 2.3 2.4 2.5 2.6 2.7 2.8 2.9 3.1

An example of ad-hoc network .............. An expanding ring search algorithm.............................................................. Applications of ad-hoc network ............. A MANET with three clusters ....... Source node propagate the RREQ packet .. TTL- Sequence based ERS Node A sends the RREQ message to its intermediate node ..

2 3 4 11 13 15 17

B relay the RREQ message .... 17 D and C broadcast the message ..... Node B and node A relay the message ...... Node B and node A relay the message .. 18 18 18

DTC: Two cases (a) and (b) ... 21 An example of two hop based method 32

Chapter 1
Introduction
1.1 Overview

Mobile ad hoc networks are wireless networks, it consist a collection of wireless nodes that are communicating to each other directly and indirectly. There is a source node, intermediate node, and destination node in the mobile ad hoc network for communication. Intermediate nodes are relay nodes they work as a router. The intermediate nodes need to relay the packet to other intermediate nodes. Source node initializes a communication to destination. Wireless mobile ad hoc networks have two categories that are infrastructure based and infrastructure less. In infrastructure based wireless network wireless nodes are communicating to each other by a coordinator that is Access points. But there is no need to any central coordinator in infrastructure less networks. With respect to their routing nature routing protocols are two types; reactive and proactive protocols. We can differentiate these two categories on the basis of the calculation of routes. We can call Reactive routing protocols as on demand protocols. Source initiates the route discovery process and flooding the packet in the entire network. So its take more energy and take more time also create the overhead across the network. To improve the route discovery process, enhance the time delay and energy consumption and stopping the unnecessary flooding in the network. In normal flooding route request propagate in the network up to TTL value goes to 0. So normal flooding creates the broadcast storm problem [1]. Normal flooding consume more bandwidth more energy and more time delay it can be reduce to stopping the flooding of packets in the entire network.

Chapter-1

Introduction

Figure 1.1: An example of mobile ad hoc network [1]

Figure 1 shows the example of mobile ad hoc networks in which source node send the message for finding the destination. Several reactive routing protocols for MANET, like DSR and AODV flood the ROUTE REQUEST messages to discover routes. In a controlled manner a ROUTE REQUEST RREQ packet is flooded in the entire network. The request RREQ is respond by a reply that is ROUTE REPLY packet which comes from the destination node or an intermediate node that has a route to the destination node. ROUTE REQUEST messages typically contain a TTL value which introduces how many numbers of hops a particular ROUTE REQUEST will cover. Suppose the source node sends the message with the TTL value 5 than the message live up to 5 hops beyond that the packet will not transmit. The expanding ring search in figure 1.2 shows how to send RREQ packet. In this figure source node sends the RREQ packet to its five intermediate nodes. These intermediate nodes are 1 hop distance from the source node.

Dept. of CEA, GLAU, Mathura

Chapter-1

Introduction

Figure 1.2: An expanding ring search algorithm

Every node receives the packet and checks the RREQ packet and extracts the information about the destination after that check in its route cache if there is any information about the destination node then reply sends to source node. Otherwise increase the hop count by 1 and further send the RREQ packet.

1.2

Motivation

Mobile ad hoc network is a field in which secure routing is the most popular areas of the research. For mobile ad hoc routing, designing a protocol which consumes minimum energy is a challenging task because mobile ad hoc network has some characteristics such as, insecure operational environment, lack of central authority, frequent topology changes, shared radio Channel, rapid node mobility and limited availability of resources. There are many number of strategies proposed for control the flooding in the network. The main focus on reactive routing protocols, because the flooding problem comes in reactive protocols like as in DSR and AODV. Reactive routing protocols have their own limitation because there is blind search for destination in the network so source node flood the packet in the entire network in order to find the destination so protocol have a challenge to use minimum energy and take minimum time to find

Dept. of CEA, GLAU, Mathura

Chapter-1

Introduction

the destination. In this thesis, we proposed an efficient route discovery process which takes minimum time delay. In market side if we see wireless ad hoc network then in mobile computing it mostly used where nodes are mobile in nature. It also used in computers and laptops so mobile communicate with other mobile devices by wireless connection and we can connect the computer with other computer by wireless medium for communication. The application of mobile ad hoc network in military field, in which there is need to transmit the data in secure fashion, in emergency services there is need of mobile ad hoc network like in hospitals or disasters areas where we need to exchange the information for relief, we have to need wireless ad hoc network to exchange the information in robot acquisition system, we have to need for wireless ad hoc network.

Figure 1.3: An application of mobile ad hoc network

In field of business suppose there is any meeting held in any company so by using ad hoc networks we can send the information from one person to all the person those are in the conferences hall for meeting without any wired connection and we can also use web camera for face to face communication, in the field of entertainment we used wireless ad hoc network to share the games, music, videos, data etc., in personal area networks, where one person exchange the information to other person

Dept. of CEA, GLAU, Mathura

Chapter-1

Introduction

without any external medium the information is exchange by body surface because device take the body surface as a medium, to complete our daily necessities home networking we used in wireless ad hoc network to complete our daily necessities, in field of education we used wireless ad hoc network to collect the various types of study materials, in sensor network wireless ad hoc network is used as a infrastructure in which sensors are there and we can exchange the information with use of the sensors, in surveillance system we used wireless ad hoc network to exchange the important information.

1.3

Issues and Challenges

One of the most complicated problem in mobile ad hoc networks is routing which have a seamless connectivity to other devices. There is no default route is available because of multi hop routing. Every node behaves as a router or as a relay node and forward packets to each others and sharing the information between mobile nodes. So there are many challenges of routing protocol 1.3.1 Limited battery life

The devices have limited battery power, due to limited battery life time, establishes the nodes in the mobile ad hoc network. QoS should consider the battery power and it contains calculation of battery consumption corresponding to resource utilization. Each node has a limited battery power so maintain the each node battery level is a challenge for mobile ad hoc networks. 1.3.2 Mobility of the nodes

In mobile ad hoc network topology is dynamically changed, for mobile ad hoc network it is challenging to keep the connection between the nodes when the topology changes. Due to the mobility sometimes the connection between the nodes frequently disconnected so when they reconnect the network it takes more energy. The system and application and also network are not aware when the topology frequently changed. 1.3.3 Lack of central corrdination

In Wireless ad hoc network nodes are free to connect with other node because it is infrastructure less. There is no central node. Nodes are distributed in the network. So challenges of this, it creates more overhead to exchange control information and consume high bandwidth by nodes.

Dept. of CEA, GLAU, Mathura

Chapter-1

Introduction

1.3.4

Energy Balancing

In mobile ad hoc network energy balancing is big issue because all the packets in routing protocols are sends at maximum power level. So challenge in energy balancing that how we balancing the energy to send the packets by which we can save energy. 1.3.5 Real time traffic support

Reservation of resources is very complicated in wireless ad hoc network because it has no infrastructure because of frequent disconnection or partitions of the network. Where nodes are mobile in nature it could not support for real time traffic. 1.3.6 Bandwidth constraint

There is abundant bandwidth available in wired network. But the radio band is limited in wireless network. In the mobile ad hoc networks when there are more overhead across the network it takes more bandwidth. Routing protocols have a constraint that is imposed by limited bandwidth to maintain the topological information. 1.3.6 Resource constraint

Battery and processing power are the two limited and essential resources that create the major constraint for the nodes in ad hoc wireless networks. Node makes bulky and less portable when the battery power and processing ability is increased. 1.3.7 Error-Prone Shared Broadcast Channel A node is received packet by all nodes which are in the direct transmission range due to nature of broadcast transmission. A node transmission do not affect any on-going session should get access to the shared medium. Since multiple nodes contend for same channel simultaneously then occurrence of collision is quite high. To minimize collision MAC Protocol should ensure that all nodes are treated. 1.3.7 Energy Conservation

Mobile ad hoc networks are without a fixed infrastructure. It has to rely on portable, limited power sources. A node has to forward messages for other nodes in the same network in an ad hoc network. The most important complication in ad hoc networks is the issue of energy consumption. From the networking point of view, develop better batteries and hardware with lower power consumption. One of the solutions for saving energy in mobile ad hoc networks is the reduction of radio transceiver power, which is often the single largest consumer of power.

Dept. of CEA, GLAU, Mathura

Chapter-1

Introduction

1.3.8 Scalability When a large number of nodes presences in the network scalability are defined, an acceptable level of service it is able to provide. Scalability is one of the most important issues of ad hoc networks.

1.4

Objective

In the mobile ad hoc network, communication can be stabiles between source and destination by different routing protocols. One is proactive routing protocols in which each node maintains the table which contains the information about every node in the network. But the proactive routing creates more overhead across the network because each node maintains the table. It also consume more energy because all time node is in active mode. The second one is reactive routing protocols in which table not contained by any node. In reactive routing protocols source node flood packet in the entire network in order to find the destination it is a blind search in which we dont know where is destination. In the reactive routing protocols expanding ring search algorithm is used to flood the packet in the entire network so it takes more time and more energy. According to proposed methodology we can reduce the time delay.

1.5

Contribution of the Thesis

This thesis is written to report the method that control the flooding of packets in the network by using hop based method. Some major contributions of our work are pointed as follows: First sends the RREQs in the network Each node has information about 2 hop neighbor nodes. Flooding more controlled by our approach is more accurate and efficient than blocking ring search algorithm.

1.6

Outline of Thesis

In this chapter, we discuss an outline of our problem and also explain overview of the problem, motivation of problem which define why we take this problem, objectives of the problem in which sequence we take a problem, define the contribution of the thesis, and also we describe its base how we reach to our approach. This chapter is

Dept. of CEA, GLAU, Mathura

Chapter-1

Introduction

organized in a way that it includes a brief description in which order the contributions made, and it also concludes steps in which way we bind up our thesis. Chapter 2: Literature Review This chapter provides clear background knowledge for this research study area. It illustrates the recent research studies on reactive routing protocols and reviews some related previous work. Chapter 3: Methodology In this chapter we discuss the complete methodology of our proposed work and it also briefly explains the approach. In this chapter firstly we give the little introduction about problem then define the proposed approach step by step how we flood the packet in the entire network in reactive routing protocol using expanding ring search algorithm. Chapter 4: Results and Discussion In this chapter we discussed the complete methodology of our proposed work and it also briefly explains the approach. In this chapter firstly we give the little introduction about problem then define the proposed approach step by step. How we reduce the time delay of the ring in the routing protocols using two hops based expanding ring search algorithm. Chapter 5: Conclusion and Future Directions This chapter of the thesis gives the final outcome of the work done and also justifies that how the proposed work is significant for further study to move ahead in the direction of mobile ad hoc network.

Dept. of CEA, GLAU, Mathura

Chapter 2
Literature review
2.1 Introduction Wireless ad hoc Network is infrastructure less where nodes communicate with each other directly or indirectly in wireless network and each node act as a router. We used wireless ad hoc network in many applications like military, emergency services, mobile computing, home networking, disaster areas, surveillance system, for deliver the information or deliver local information etc. Mobile ad hoc network prefer to the set of networks where all nodes in the network have equal status and free to communicate with each other in specific range. Mobile ad hoc network is mobile, not static in nature. So it is suitable for many applications where central node is not relied. By using expanding ring search algorithm in the reactive routing protocols, we can improve the more energy consumptions very efficiently. Expanding ring search algorithm can apply on any reactive routing protocols like AODV and DSR. This algorithm is basically a concept how to stabilize a path from source to destination. So there are many proposed scheme which are given by different author. This algorithm is applied on any protocol to find the path which consume minimum energy and take minimum time delay also take minimum overhead, so our goal is to tackle these issues. Energy and time delay is important issue in mobile ad hoc network. So there are many strategies defined in this chapter some strategy consume minimum energy but take more time and more overhead, some energy consume minimum time delay etc. This chapter is organized such as section 2.2 describes the various approaches, in section 2.3 define the comparative analysis of various approaches and section 2.4 finally summarizes the whole concept.

Chapter 2

literature review

2.2

Expanding Ring Search Approaches

Broadcast storm problem [1] is very serious problem in mobile ad hoc network. This is caused by flooding of packet. The drawback of flooding in the mobile ad hoc networks is redundant broadcasts, contention, and collision. In the redundant broadcast, one node broadcast the same message more than one time. So it creates overhead across the network. In contention, suppose source node sends the message to its intermediate node then each intermediate node which receive the packet to the source node, contend to each other to broadcast further. Collision also creates problem because of lack of RTS/CTS dialogue. Simple way to find out the destination, flooding the entire network but it create problems and take more energy, take more time, create more overhead in the network. It creates the contention collision and redundant rebroadcasting. So there are some mechanisms to reduce these problems. One of them is probabilistic mechanism, when intermediate node receives the packet to other node it rebroadcast the packet by probability P. When P=1 its sends packets in broadcast manner. Another mechanism is counter based; when a message rebroadcast by a host than message may be blocked by back off procedure or by busy medium. So host has a chance to hear same message repeatedly from the other host before the host actually start their transmission. So author used counter based mechanism in which c is a counter that is initialize by 1 when the message heard first time and another C is a threshold. If message heard more than one times than increase c by 1 and check c<C, if it is true than resume the interrupt and if c=C than cancel the transmission. So counter contains how many times the number of time packet received. In distance based mechanism; author used the relative distance to decide whether to broadcast the message or not. If A sends message to B, if the distance between A and B is 0 than B does not receive the message from A. If distance between them is greater than 0 than broadcast the message up to that distance. Distance shows the coverage area. It reduces rebroadcasting of the messages. Here also a threshold that is D. Initializes dmin, when message broadcast first time .If dmin is less than D than cancel the transmission of message otherwise resumes the interrupted waiting. Message waits for a random number of time slots and after that submits for transmission until the actual transmission will not start.

Dept. of CEA, GLAU, Mathura

10

Chapter 2

literature review

Another one is location based; if we acquire the location of the destination node than we can minimize the flooding problem. Such an approach that is GPS (Global position system) receiver device that finds the location of the destination. It contains the location information in the form of x y-coordinates. It also provides the location in the form of 3-D which is longitude, latitude and altitude. Author defines a threshold A and compares the threshold value with the additional coverage AC. If A<AC rebroadcast the message otherwise not. Another one is cluster based; cluster formation based algorithm is used to find the cluster head.

Figure 2.1: A MANET with three clusters

A node periodically sends the beacon packets to show its presence in the network. A node that have minimal local ID selected as a cluster head and each cluster head has some member nodes and the gateway node that pass the information from one cluster to another cluster. So when a node broadcast a message to the other node, if receiving node not a gateway node than stop the broadcasting of packet. After that it applies one of the above conditions to decide whether to rebroadcast packet or not. The Strength of this paper is that it reduces the overhead in the network and also it gives the improvement over the basic flooding. Limitation of this paper is that proposed scheme is not suitable for the entire problem.

Dept. of CEA, GLAU, Mathura

11

Chapter 2

literature review

Flooding strategy for target discovery [2] in wireless network, which takes the problem to find the best flooding strategy that requires minimum time delay and take minimum energy. In this paper author takes two approaches that either flood the entire network or flood that follow ring wise mechanism. Which of the two is better if it apply ring mechanism than how many rings are suitable to flood the network to find the destination? So author takes two concepts that are two ring and three ring mechanisms. In this paper author takes two category that are large scale network and other one is small scale network. In large scale network, it ignores the edge effect, while for small scale network takes edge effects. In large scale network takes flooding strategy problem, and take hop limit strategy to control the flooding in the network. Source node sends the RREQ packet by setting the hop limit value if it fails to find the destination that it increase the hop limit. In small scale network, source node sends the packet in the entire network when network is small. Here edge effect taken into account. If the source node knows the distance of the boarder node than it can adjust the hop limit according to that distance. Author takes two ring schemes for this case. Other is self location aware; if node does not have information about the border node it simply flood the entire network. So overall, two ring scheme reduce the cost of small scale network. In two ring scheme, set the hop limit firstly up to [m/2] and the second time take the hop limit up to m. So it reduces the cost of the network. If it takes three ring scheme , set the hop limit first time 1 and second time it takes the hop limit up to [m+1/2] and in third time take hop limit m. author takes DSR and take the hop limit for that 1 in first attempt and in second attempt it takes M. In multi target discovery, it searches for multiple destination or target but in single target discovery, it finds the one target. So here we take multiple ring searches to find the target node. Here the strategy one-out-of-m is used. Optimal L [3] value also reduces the time delay and energy consumption. Here it examine the behavior of L after that examine the search strategy of L. DSR and AODV are the reactive routing protocols they are not operate until route from source to destination are found. Route discovery process starts on route demand from source. There are many techniques proposed to enhance the route discovery process. ERS algorithm is one of the techniques used to find the route from source to destination [5]. ERS algorithm support reactive routing protocols such as AODV and DSR.

Dept. of CEA, GLAU, Mathura

12

Chapter 2

literature review

In ERS [5] there is multiple rings route discovery process which starts from the source node. Source node sends the RREQ packet to the intermediate node that is one hop distance from the source node. Intermediate node receives the RREQ packet and check in its route cache. If there is information about destination node then sends reply back to the source node by RREP packet otherwise forward the packet to the next ring. Again process will start of searching of the destination node from the source node until destination node will found. Blocking ring search [6] approach is an alternative ERS approach. BERS support the reactive routing protocols. In the expanding ring search algorithm when source node sends the RREQ packet to the one hop neighbor node. When the intermediate node receive the RREQ than each node check in its route cache if there is any information about the destination than node send the reply packet RREP back to the source node if not than further broadcast the packet to the next hop neighbor node until the destination node found . After receiving the RREP packet the source node sends the chase packet with hop count to stop the further broadcasting because of that this strategy called the blocking of ERS.
C S A D

Figure 2.2: Source node propagate the RREQ packet [5]

In figure 2.2 source node sends the RREQ packet to its neighbor node. When source node sends the packet it also sends the hop number with that packet. When the source node sends the RREQ packet with the hop number 1 than only 1 hop neighbor node receive the packet after that each node check in its route cache. If there is any information about the destination node than reply sends to the source node. If none of them have information about destination node than increase the hop count by 1 and again send the RREQ packet.

Dept. of CEA, GLAU, Mathura

13

Chapter 2

literature review

The main purpose of the expanding ring search algorithm is to find the node that have information about destination node so node flood the packet in the entire network that create some problem such as energy consumption, create overhead or take more time. To control the flooding, author proposed TTL sequence-based expanding ring search algorithm. In TTL sequence-based expanding ring search, sets a TTL value before broadcast the RREQ packet, after broadcast the RREQ packet with TTL value, message will send up to TTL value becomes 0. Within the TTL value if the route for destination node found than stop the broadcasting of RREQ packet otherwise again send the RREQ packet from the source node with the increased TTL value. So the problem of this method is that, it creates more overhead at the source because if destination node does not found for specified TTL value than it repeatedly starts broadcasting from the source node by increasing the TTL value. So it consumes more energy also and takes more time than the basic route discovery approach. So in this paper author propose an efficient algorithm that consumes minimum energy than the TTL sequence-based expanding ring search. In the blocking expanding ring search algorithm there is a newly control packet that is stop instruction and also a hop count which reduce the consumption of energy. Both methods are same in their working strategies. But one of the differences between them is, in blocking expanding ring search algorithm, when the destination node does not found than it does not repeatedly start from the source node. It does rebroadcasting of the packet from that intermediate node which has no information about the destination node. So on behalf of source node the intermediate node behaves like a relay node. In figure 2.3 the source node broadcast the RREQ packet with the hop count 1, so the intermediate node which is one hop distance from the source node receive the RREQ packet and check in its route cache. If there is information about destination node than stop the flooding of RREQ packets in the network and send the RREP packet to source node. If no information is available in the first ring than the node increases the hop count by 1 and now the hop count is H=2 than rebroadcast the packet. This process repeated until route to the destination node does not found. Each node in the blocking expanding ring search algorithm wait for 2H unit time which is need to decide that the RREQ packet broadcast further or not.

Dept. of CEA, GLAU, Mathura

14

Chapter 2

literature review

Source node

Intermediate nodes

Route node

1st broadcasting

2nd broadcasting 3rd broadcasting

1st ring H=1

2nd ring H=2

3rdring H=3

Figure 2.3: TTL- Sequence based ERS So 2nd ring receive the packet and wait for 2H time because it has no information about destination node. The 3rd ring has the information about the destination node than it sends the hop count with the RREP packet. So the source node sends the stop instruction, after receiving the RREP from the destination node, with the hop count to all the nodes and stops the flooding of RREQ packet in the entire network. Within the waiting time the intermediate node receive stop instruction than intermediate nodes do not broadcast the RREQ packet otherwise increase the hop count by 1 and broadcast the packet. So by using this approach, it reduces the flooding of RREQ packet in the entire network. Energy consumption in BERS depends on the number of nodes in a ring. In ring 1 there is only one node that is source node so the power consume by that node is 1 unit. In 2nd ring there are n1 number of nodes so energy consume by them are n1 unit. So the total energy consumed in BERS is
st

(2.1)

Dept. of CEA, GLAU, Mathura

15

Chapter 2

literature review

(2.2)

The difference between the

and

is

(2.3)

Strength of BERS improves the energy consumption in route discovery. It depends on the number of nodes involved in the flooding and number of rings. And the Limitation BERS takes more time then TTL- based ERS because each node waits for 2H time before broadcast the packet to other nodes. BERS* (Enhanced blocking expanding ring search) [7] is based on ERS algorithm. It is energy efficient but more time efficient. It takes half amount of time than BERS. In BERS each node wait for 2H time before propagate the packets but in BERS*, intermediate node takes h unit time. It minimize the time latency approximately half than the BERS. In EERS, [8] when node sends packet to other nodes, if the node receive the packet first time it will takes the message and propagate to other nodes, if node receive the packet more than one time it will discards the packet without saving the information such as node ID about the neighbor node or source node. But in EERS, when intermediate node receives the packet second time than before discarding the packet it will take the information about the neighbor node. In EERS some nodes in active state and other nodes in rest state. The nodes which are in active state relay the RREQ packet and other nodes do not relay the message. Every node contains the variable which is called relay. Relays value shows that the node is in active state or not. If the value of relay variable is true than nodes relay the packet otherwise nodes do not relay the message it will be in silent state. The incoming messages decide the

Dept. of CEA, GLAU, Mathura

16

Chapter 2

literature review

value of relay variable because the incoming message contains the extra field which is the predecessor. Every packet in the network contains the predecessor value. Figure 2.4 shows the working of enhanced expanding ring search algorithm [8]. Suppose A sends the message to B and than A send the predecessor value in that message. Each node maintains two information, first one is relay information and second, first time forwarded message. suppose A is source node it sends the packet to intermediate node B and sends own address as a predecessor to B at that time the value of relay will be false and value of FTF be true. Because node A first time sends the packet to other intermediate node. After receiving the message Node B compare its own address with predecessor value if they matches than node relay the message to other node if not than does not relay the message and reach in silent mode
Relay= false FTF= true Relay= false FTF= true B A C Relay= false FTF= true Relay= false FTF= true D Relay = false FTF= true

Figure 2.4: node A sends the RREQ message to its intermediate node

Relay= false FTF= true

Relay= false FTF= true B A Relay= false FTF= true Relay= false FTF= true C D Relay = false FTF= true

Figure 2.5: B relay the RREQ message

Dept. of CEA, GLAU, Mathura

17

Chapter 2

literature review

Figure 2.5 shows relaying of node B [8]. Now B sends the message to C than A is predecessor for C because the message from B contains the address of node A as predecessor than check Cs address with A. The value of predecessor and Cs address not matched so the node C does not relay the packet to other nodes. But node B also send the message to node A so the value of predecessor match with As address. Node A will relay the packet. Figure 2.6 show that the node C and node D broadcast the message at first time. When node C sends the message to node A, node A checks the predecessor address, because node As address match with predecessor so node C will relay the message to other nodes. At the same time D also sends the RREQ packet to B and C. Because of node Bs address match with the predecessor value that is node Bs own address so node B relay the message
Relay= false FTF= true

B A Relay= false FTF= true Relay= false FTF= true C D Relay = false FTF= true

Figure 2.6: D and C broadcast the message

Relay= true FTF= true B A Relay= true FTF= true C Relay= false FTF= true D Relay = false FTF= true

Relay= false FTF= true

Figure 2.7: node B and node A relay the message

Dept. of CEA, GLAU, Mathura

18

Chapter 2

literature review

At the end, only node B and node C relay the packet and other nodes will be in silent state. Node W, C, D conserves energy. Fig. 2.7 shows that only node B and node A relay value is true. There are two phases in which first one collecting neighbors information and second is reducing the overhead of flooding [8]. In the first phase each node sends the message to all the nodes without taking the relay variable. Relay is set to true and set to false according to the node state. So in second phase node relay the message. After the first phase some nodes has the value of relay as false and some has value as true. If node has value as false not relay the packet otherwise relay the packet. According to the author, the proposed methodology creates problem for the border nodes relay value because in mobile ad hoc network nodes are not fixed they move from one position to other position so the border nodes relay value never changed and the TTL value for the border node is always 0 so they never retransmit the message. So to remove this problem author reset the value of relay variable after T seconds. The Strength of EERS, some node relay the message and other will be silent so energy in EERS efficiently used. And Limitation of EERS Additional field predecessor is used so it uses more bandwidth. In AODV based ERS [9] algorithm, a hop-based prediction Scheme is used in which hop is predicted for the destination on the basis good TTL value. Good TTL value is select on the basis of prediction of the destination node. Based on the previous hop number or the destination number fined the new destination. This scheme takes the source node, destination node, hop count to the destination node and find out the mean of previous hop count of some destination. The mean value gives the value of hop count which is destination hop count. There is a hop table in which the information about destination node, number of records to some destination, mean value of hop to some destination, hop count of latest route request and time of latest record are available. So it uses the hop table to predict the destination and reduces the time latency. When source node sends the RREQ packet to one hop neighbor nodes. They receive the RREQ packet and check the redundancy of the packet. If the RREQ packet is duplicate than discard the RREQ otherwise search the information in hop

Dept. of CEA, GLAU, Mathura

19

Chapter 2

literature review

table of the destination. If there is an item in the destination hop table than store the hop count, than calculate the mean value of hop to the destination and some number of records of destination. If there is no item in the destination hop table than create an item also store the hop count and again calculate the destination node and mean value of hop to the destination and also calculate some number of records of destination. These steps performed by every node in the network. So after that data to be send to the destination search in the route table if there is a valid route to the destination sends the data otherwise searches in hop table. If there is an item to the destination node than calculate the hop count and set the TTL value to the hop count and send the data. If there is no item in hop table than flooding the packet in the entire network. AODV based HP-ERS algorithm reduces the end to end delay by reducing the network congestion and also improve the routing overhead by reducing the number of RREQ packets. Algorithm saves some resources such as bandwidth which is important point in mobile ad hoc network and also save power. This approach reduces the network overhead but on the other side it requires management of the hop table. DTC(Dual tier channel) [10] have two tier approach, based on virtual channel that is not actual channel. Two channels are used here, that takes two different speeds. One speed is slow and other is fast means first one slower than the second one. In the network no node waits before broadcasting the message to the other node. Every node broadcast the message without waiting so many times. RREQ packet broadcast by the non route node in the network, also beyond the network because no node waits for chase packet. In order to catch the RREQ, chase packet is being sent in the fast speed than the RREQ speed. A threshold is predefined to control both of the speeds. In DTC there is a need to the intermediate nodes to find there geographical location in the form of hop distance. If the hop distance is less than the threshold value than nodes act as ERS. If the hop distance is greater than the threshold value or beyond the threshold value than nodes will acts as BERS. The speed of RREQ will slow down as soon as route node is found. Hc and Ht term is used in DTC. Hc, the ring in which the RREQ packet caught and Ht the ring where the speed is changed from v1 to v2. v1 speed is faster than speed v2. Source node broadcast the RREQ packet at the v1 speed up to threshold value that is ht. If route node or destination node found the speed is converted into v2

Dept. of CEA, GLAU, Mathura

20

Chapter 2

literature review

so easily catch when sending the chase packet, chase packet speed is also v1 to catch the flooding. The ring where the route node is found is Hr ring. In figure[10] DTC have two cases in the first case the route node is found after the threshold value so the propagation speed will converted into slower speed after the threshold value, in the second case route node is found before the threshold value so up to threshold value the propagation speed will not changed. So the first case shows Ht<Hr, and second case shows that Ht>Hr.

Figure 2.9: DTC: Two cases (a) and (b) Hc be the ring, where we expected that the RREQ packet to be caught by chase packet. Author calculates the Hc. Case 1:

(2.4)

The left hand side of this equation represents the RREQ propagation time and chase packet time in v1. The right hand side of this equation represents the RREQs continues propagates in v2(h). After solving the equation author have, =2 Case 2: +
(2.5)

(2.6)

Dept. of CEA, GLAU, Mathura

21

Chapter 2

literature review

(2.7)

After solving the equation When Ht = Hr, we have Hc = 3Hr Energy consumption in DTC: Author consider two scenario In the first scenario, There is a ring i in which ni number of nodes that are the broadcasting nodes, where i = 1,2,, Hc. Energy consume by DTC is calculated as
(2.8)

(2.9)

The second scenario is considered by the author as, nodes are distributed randomly and uniformly in the geographic area. = density of the nodes r = Radius of the imaginative circle center by the source node in the ring I, the

number of nodes

(2.10)

when Hr = Ht

)
(2.11)

Latency in DTC: Author consider here also two scenario Case 1:

Dept. of CEA, GLAU, Mathura

22

Chapter 2

literature review

(2.12)

After solving this equation, we have

(2.13)

Case 2:

(2.14)

(2.15)

For a particular case, Ht=Hr

(2.16)

So author concluded that, Compare both the strategy BERS and DTC. DTC reduce the time latency but it takes more energy than the BERS. It is suitable for small network because when we use DTC strategy in large network there is more chance that RREQ Broadcasts beyond the network. In BERS each node wait for 2H time, after 2H time node broadcast the message so its take more time but reduce the energy consumption.

Dept. of CEA, GLAU, Mathura

23

Chapter 2

literature review

TABLE 2.1: Comparative Analysis

Factors/ Papers 1 2 3 4 5 6
Broadcast strom problem Flooding strategy for target discovery Optimal L value The expanding ring search Blocking expanding ring search Enhanced blocking expanding ring search Limiting Flooding expenses En efficient expanding ring search AODV based HP-ERS algorithm Dual tier channel YES YES YES YES NO

Methodology

Energy consumption

Time Delay YES YES NO YES YES

Overhead

NO NO NO YES NO

YES

YES

NO

NO

YES

NO

8 9 10

YES

NO YES YES

NO NO NO

YES NO

If we compare the proposed method to the existing approaches then our approach is reducing time delay so our methodology contains two hops information at every node so we can get the destination information quickly.

2.3

Summary

In this chapter we discuss the various approaches for path finding which consume minimum energy minimum time delay and also take minimum overhead, by using these we know the previous and recent work about the path finding algorithm which existing in reactive routing protocol. Using some of those approaches we can reduce the some problem like energy consumption and time delay e.t.c. We also analytically

Dept. of CEA, GLAU, Mathura

24

Chapter 2

literature review

analysis comparisons based on some parameters like packet delivery ratio, time delay based on these we can conclude various methodology.

Dept. of CEA, GLAU, Mathura

25

Chapter 3
Proposed Methodology
3.1 Introduction

A mobile ad hoc network consists of a collection of wireless nodes that communicate to each other directly and indirectly. There is a source node, intermediate node, and destination node. Intermediate nodes are relay nodes. They work as a router. The intermediate nodes need to forward the packet to other intermediate nodes. Source node initializes a communication to destination. Wireless mobile ad hoc networks have two categories that are infrastructure based and infrastructure less. In infrastructure based wireless network wireless nodes communicate to each other by a coordinator that is Access point. But there is no need to any central coordinator in infrastructure less network. In mobile ad hoc network any device is free to move anywhere because the nature of device is mobile. Mobile ad hoc networks devices are such as laptops mobiles, PDAs .Mobile Ad-hoc Networks are the classical ad hoc network which is used in the military sector. By using the mobile ad hoc networks connect the tanks, planesand soldiers at the battlefield. Mobile ad hoc networks are completely selforganizing. The other areas where the mobile ad hoc networks applied are disaster area like tsunami for recovery and also for relief. The other area where the mobile ad hoc networks used is in communication, home networking, and education, medical, to exchange the local information, robotics, ubiquitous computing, wearable computing, sensor network, and entertainment and in field of business. In mobile ad hoc networks reactive routing protocols are not table driven protocols. It is on demand routing protocols. In the reactive routing protocols route is established only when source node wants. If there is no route available for destination than source node needs to send the RREQ packet to all the nodes in broadcast manner.A route request carries the source ID, the destination ID, the source sequence number, the destination sequence number, the TTL value and broadcast ID. So at each

Chapter 3

Methodology

node destination address checked if it matches with that node than destination sends the RREP packet. TABLE 3.1: Information of RREQ packet Src ID Dest ID SrcSeqNum DestSeqNum TTL value Brdcast ID

Natural flooding of RREQ packet in the network in order to find the destination, than each node receives the RREQ packet from every other node and check for destination. Mobile ad hoc networks have an issue of battery power and have a limited time. So the natural flooding of packets in the network consumes more energy because each node receives the packet so it takes battery power and also takes time delay. To solve this problem blockingexpanding ring search came into existence, which works based on onlyby sending the RREQ packets ring by ring. Ring is formed by hop count means that the first ring is formed by those node which are 1 hop distance from the source node and second ring is formed by those node which are two hop distance from the source node and so on. So this process controls the flooding in order to save the battery power but not time. And other one is enhanced expanding ring search algorithm came into existence, in which nodes take half amount of time than the blocking expanding ring search algorithm. So this process reduces the time delay. In order to solve the problem of both blocking expanding ring search algorithm and enhanced blocking expanding ring search algorithm, our work attempts to control the flooding in the network which takes minimum time. In blocking expanding ring search algorithm when source node sends the RREQ packets in the network than each node which is one hop distance from the source node, receive the packet and each node has its own route cache in which they contains some information about neighbor nodes and destination node. So every node after receiving the RREQ packet check its address in its route cache if there is any match with the message address than destination node found and reply sends to source node. In blocking ring search algorithm every node before sending the RREQ packet wait for 2H time, if node receive the stop instruction in that time than does not broadcast the packet to the other nodes. If it does not receive the stop message than forward the message to other nodes. So it takesmore time than the natural flooding mechanism. In

Dept. of CEA, GLAU, Mathura

27

Chapter 3

Methodology

enhanced blocking ring search algorithm each node wait for H time before broadcast the RREQ packet to other nodes. So it takes minimum time than the blocking ring search algorithm but consumes more power. Before proceeding, we introduce some of the key contribution of our approach: First source node sends the RREQ packets to all neighbor nodes by. Each node has information about 2 hop neighbor nodes. Flooding controlled by our approach is more accurate and efficient than blocking ring search algorithm. In the proposed approach source node sends RREQ packets to all the neighbors nodes which receive the packet and wait for a unit time and then send to other nodes. Each node maintains the information about the two hop neighbor nodes. Which the time delay. In this chapter we define the brief description of the approach in Section-3.2 and in Section-3.3 finally summarize the concept at the end of this chapter.

3.2

Hop Based-BERS

In normal flooding, source node sends the RREQ packet to the other nodes in broadcast manner so every node receives the packet to every other node. Every node receives the same message more than one times so it creates more overhead across the nodes.Due to this battery of nodes are depleted. In BERS algorithm source node sends the RREQ packet to each node that all are one hop distance from the source node. Intermediate nodes receive the RREQ packet and check in its route cache. If the intermediate node have information about the destination node than that intermediate node sends RREP to the source node.If it does not, then send RREQ packet to next intermediate node. Each node waits for 2H time before sending the RREQ packet. Each node communicates to other node directly in multi ring ERS algorithm. To take the advantage of BERS algorithm we proposed the 2-hop based ERS scheme. In this method every node contains the information about 2 hop neighbor nodes. When source node sends RREQ packet to intermediate nodes that are 1 hop distance away from the source node. Each node contain a table, if any information in its table about the destination node are available than it sends reply to the source node to stop the flooding. Table 3.1 shows the algorithm of source node, Table 3.2 shows Dept. of CEA, GLAU, Mathura 28

Chapter 3

Methodology

the algorithm for intermediate node, Table 3.3 shows the algorithm for destination andTable 3.4 shows the blocking procedure. Algorithm 1 covers the source node action for the discovery of the route. First itsends the RREQ packet to the one hop neighbor node. The other work of source node is to broadcast the chase packet or stop instruction after the RREP is being received. And use the RREP packet which receive first and drop any later packet.

TABLE 3.2:Algorithm for source node Algorithm 1 for Source node 1. With H = 1 and max H Broadcast RREQ packet. 2. Source node wait for RREP until RREP received 3. The stop instructionand H is broadcast to all the nodes within the ring.

Algorithm 2 shows the action of intermediate node. The work of intermediate node is to forward the RREQ packet to other intermediate nodes. It also receives the RREP packet to forward to the source node and also receive the stop message. After receiving the RREQ packet node check the H (hop count), if it is bigger than the max value than drop the packet otherwise check in its route table.If there is information about the destination node than send the RREP to the source node with the hop count value.If there is no information about the destination node than wait for 2H time after that forward the packet to the other nodes. At the waiting time if node receives the stop instruction than stop the broadcasting of the packet and then erase the pair of destination and source node from hop table. At the waiting time if no stop instruction itreceive than increase the hop count value by 1 and rebroadcast the RREQ packet. This process repeated until the destination node will found.

Dept. of CEA, GLAU, Mathura

29

Chapter 3 TABLE 3.3: Algorithm for intermediate node . Algorithm 2 for Intermediate node 1. Receive RREQ 2. After receiving the RREQ check the max j 3. If the H > max j then

Methodology

4. leave the RREQ 5. else 6. After receiving the RREQ check the hop table(which contains the 2 hop neighbor node information) 7. If the route information is in the hop table then

8. Send H and RREP to the source node 9. Else 10. Every node waits up to waiting time ( hop number) 11. While waiting do 12. Stop instruction received 13. Execute the blocking procedure 14. Erase the information of the route cache 15. else if RREP received then 16. forward to the source node 17. end if 18. end while 19. at the waiting time no stop instruction receive then 20. hop serial increase by 1 and rebroadcast RREQ 21. end if 22. end if

In algorithm 3, describe the action of route node i.e. a route node is that node which contains the information about the destination node. After receiving the 1st RREQ packet the route node sends the RREP packet with the hop count to the source node.

Dept. of CEA, GLAU, Mathura

30

Chapter 3

Methodology

TABLE 3.4: Algorithm for destination node Algorithm 3 Route or destination node 1. Wait for Receive a RREQ then 2. Send the RREP and the H to the source route (contained in the RREQ Packet) 3. end if

In algorithm 4, it shows the performance of the blocking procedure. END instruction is broadcast to the nodes to block the flooding of packet. After receiving the RREP packet and the hop count the source node sends the END instruction with that hop count Hr.So each node compare H with Hr, if Hr is bigger than the value of H than forward the stop instruction further and erase the source destination pair. If the value of Hr is equal to the H than stop rebroadcasting of RREQ packet and then it will drop the stop instruction, will also erase the source destination pair from its hop table.

TABLE3.5: Procedure of Blocking Algorithm 4 Procedure of blocking 1. compare H with Hr 2. if Hr< H then

3. Stop instruction forward to all nodes and 4. 5. erase the information of the route cache else

6. Stop instruction leave by node and 7. stop rebroadcasting and 8. erase the information of the route cache 9. end if

So these four algorithms show the working of the proposed approach which gives the better result than the previous approaches. Figure 3.1 shows the example of the proposed approach.

Dept. of CEA, GLAU, Mathura

31

Chapter 3

Methodology

Destination node Route node

Source node Figure 3.1: An example of two hop based method In the above figure source node sends the RREQ packet to its neighbor nodes than its intermediate node receives the packet and check in its hop table. If there is any information about the destination node than RREP packet will send to the source node. In this figure the node which is three hops distance away from the source node has the information in its hop table. It sends the reply RREP to the source node. This node contains a route to the destination. Using this approach we can reduce the time delay by keeping the two hop neighbor node information at each node.

3.3

Summary

In this approach we defined the procedure to control flooding in the reactive routing protocols for reducing the time delay.

Dept. of CEA, GLAU, Mathura

32

Chapter 4
Results and Discussion
4.1 Introduction

In this chapter we discussed about the result analysis of the proposed HB-BERS algorithm in the different scenario and compared it with the existing ERS algorithm using AODV protocol. Comparative study of both the approach is based on by taking different number of connection, pause time, packet delivery ratio and end to end delay. The proposed approach is simulated by using the Network Simulator-2.35(NS2.35) at P4 CPU 2.53 GHz PC with 2 GB of RAM. The x-graph shows the result comparison with AODV- expanding ring search algorithm and HB-BERS algorithm. The graphical user interference of the proposed approach shown by network animator and the table shows the different situation. Finally, we analyzed the experimental result of both the approaches. This chapter is further organized as follows: we discussed quantitatively the results of proposed methodology in Section 4.2. Some snapshots of given approach is taken by the Graphical User Interference are in section 4.3 which shows the entire procedure by using Network Animator (NAM). Section 4.4 summarizes the results and analysis of the given approach.

4.2

Result Analysis

The proposed method can be performed by using the Network Simulator (NS-2.31). NS-2 is network simulator it is a event driven simulator. At UC BerkeleyNS-2 developed and it implements the protocol like AODV, DSR, DSDV, AOMDV etc. Each protocol is defined in header file (.h file) and .cc file by using C++, so it works as backend. NS-2 is based on two OTcl (Object oriented Tool Command Language)languages and C++(object oriented language), using OTcl script topology

Chapter 4

Results and Discussion

of the network is simulated and C++ is used for fast compilation. In NS2, we have two steps first step is preprocessing in which Tool Command Language (TCL) comes; the front end tool of the NS-2 is TCL which shows the simulation and the other one is NAM which is network animator. That is used for GUI presentation of the network. We can calculate the result by using the AWK script so AWK script shows the result after calculating them. There is real time simulation of the nodes which is the Advantages of using NS-2. So finally generate the result using the NS-2 simulator.

Quantitative Analysis:
According to the proposed approach it sends all the RREQ packets to the network which takes the source sequence number, destination sequence number, destination address, broadcast ID, TTL value. So every node receive the packet and check the RREQ packet and retrieve the destination address in order to check the node address with that destination address if it is matched than no need to flood more. So in our approach it needs to store the 2 hop value information in hop table at every node. We analysis this approach numerically by taking the scenario of different nodes and different pause time like 100, 200, 300, 400, 500. We calculated the packet delivery ratio, end to end delay, using AWK script. AWK script uses the information from the trace file, the race file produced by the TCL file. The protocol definitions linked by TCL file which is written in the form of C++. The following discussion is based on output of the AODV-HB-BERS and AODV- ERS which calculate the packet delivery ratio by using the AWK script at different pause time.

TABLE 4.1: packet delivery ratio at 100 pause time No. of connection 20 30 50 100 AODV-ERS 98.9312 97.9106 98.8485 99.9546 AODV-HB-BERS 99.9334 100 99.9312 99.9696

Dept. of CEA, GLAU, Mathura

34

Chapter 4

Results and Discussion

TABLE 4.2: packet delivery ratio at 200 pause time No. of connection 20 30 50 100 AODV-ERS 96.9106 94.7867 98.7739 98.8547 AODV-HB-BERS 98.7945 99.9723 99.8431 99.9384

TABLE 4.3: packet delivery ratio at 300 pause time No. of connection 20 30 50 100 AODV-ERS 95.6294 95.6737 98.8839 97.9746 AODV-HB-BERS 99.9312 99.9848 99.8898 100

TABLE 4.4: packet delivery ratio at 400 pause time No. of connection 20 30 50 100 AODV-ERS 95.8485 98.5338 96.8673 96.7329 AODV-HB-BERS 99.9845 100 99.9717 99.9057

Dept. of CEA, GLAU, Mathura

35

Chapter 4

Results and Discussion

TABLE 4.5: packet delivery ratio at 500 pause time No. of connection 20 30 50 100 AODV-ERS 98.8507 97.9296 98.9394 96.8751 AODV-HB-BERS 99.9645 99.9328 99.9579 99.9365

Here we discussed the output of the proposed approach and existing approach.wecalculated the end to end delay with different pause time using AWK script. TABLE 4.6: end to end delay at 100 pause time No. of connection 20 30 50 100 AODV-ERS .340667 .403636 1.15242 2.40616 AODV-HB-BERS .201636 .26249 .784615 1.3549

TABLE 4.7: end to end delay at 200 pause time No. of connection 20 30 50 100 AODV-ERS 0.4756 0.465468 0.7842 4.02247 AODV-HB-BERS 0.202711 0.277621 0.68109 2.2373

Dept. of CEA, GLAU, Mathura

36

Chapter 4

Results and Discussion

TABLE 4.8: end to end delay at 300 pause time No. of connection 20 30 50 100 AODV-ERS 0.421194 0.92144 0.817442 4.4405 AODV-HB-BERS 0.241949 0.4337 0.327525 2.9875

TABLE 4.9: end to end delay at 400 pause time No. of connection 20 30 50 100 AODV-ERS 0.889226 1.4918 0.707314 3.1654 AODV-HB-BERS 0.411492 0.12257 0.52834 6.9096

TABLE 4.10: end to end delay at 500 pause time No. of connection 20 30 50 100 AODV-ERS 0.92144 0.497183 0.42547 8.26522 AODV-HB-BERS 0.43899 0.3193 0.251956 5.87939

Here we have taken average value, packet delivery ratio and end to end delay. we have shown by table and we analyzed the average of end to end and packet delivery ratio.

Dept. of CEA, GLAU, Mathura

37

Chapter 4

Results and Discussion

TABLE 4.11: Average packet delivery ratio No. of connection 20 30 50 100 AODV-ERS 98.91125 95.331475 96.7904 98.8987 AODV-HB-BERS 99.95855 99.637075 99.95145 99.9645

TABLE 4.12: Average end to end delay No. of connection 20 30 50 100 AODV-ERS 1.07572075 1.436935 1.650144 4.8743 AODV-HB-BERS 0.650918 0.8496805 0.997668 2.8967

Qualitative analysis:
We have taken some simulation parameters like carrier sensing range, transmission range, antenna gain, channel frequency and data rate, which are shown in TABLE 4.13. Table 4.13: Simulation Parameter Parameters Data rate Channel frequency Antenna gain Carrier sensing range Transmission range Values 1Mbps 2.5 GHz 1m 200 m 200 m

Dept. of CEA, GLAU, Mathura

38

Chapter 4

Results and Discussion

250 200 packet delivry ratio 150 100 50 0 20 30 50 100 No. of connection AODV-HB-BERS AODV-ERS

Figure 4.1: Graph of packet delivery ratio

9 8 7 end to end delay 6 5 4 3 2 1 0 20 30 50 100 No. of connection AODV-ERS AODV-HB-BERS

Figure 4.2: Graph of end to end delay

Dept. of CEA, GLAU, Mathura

39

Chapter 4

Results and Discussion

4.3

Summary

In this chapter, we analyze the resultsat different scenarios of proposed methodology and existing one by changing the number of nodes and pause time. For some scenarios the proposed algorithm is not performing well with respect to existing algorithms. But for most of the scenarios the performance of AODV-HB-BERS is better than the AODV-ERS.

Dept. of CEA, GLAU, Mathura

40

Chapter 5
Conclusion and Future Directions
5.1 Summary and Contributions

In this thesis, we proposed HB-BERS algorithm which gives the better result than the existing approaches. It reduces the time just half. In this thesis we propose an algorithm in which every node contains the two hop neighbor node information. So by using this concept we reduce the time delay.

5.2

Future Directions

The proposed methodology works on route discovery process and its not able to reduce the overhead of the network. The proposed method also does not give the good results in some parameter like normalized routing load. In the future direction we work on to route maintenance process and also try to decrease overhead.

References
[1] L. S. Ju, E. B. Royer and C. Perkins, Scalability Study of the Ad Hoc OnDemand Distance Vector Routing Protocol, International Journals Network Management vol. 13, pp. 97-114, 2003. [2] J. Hassan and S. Jha, Optimizing Expanding Ring Search for Multi Hop Wireless Networks, IEEE Global Telecommunications Conference, pp. 25302004. [3] K. Carlberg and J. Crowcroft, Building Shared Trees Using A One-To-Many Joining Mechanism, ACM Computer Communication Review, pp. 511, 1997. [4] P. Cao, E. Cohen, K. Li and S. Shenker, Search and Replication in Unstructured Peer-To-Peer Networks, in Proceeding of the ACM Sigmetrics Conference, pp. 9-14, 2002. [5] D. Koutsonikolas, S. Das, H. Pucha and Y. C. Hu., On Optimal TTL Sequence-Based Route Discovery in Manets, in Proceeding of the 2nd ICDCS International Workshop on Wireless Ad Hoc Networking, pp. 395-399, 2005. [6] Robert and R. Samir, Query Localization Techniques for On-Demand Routing Protocols in Ad Hoc Networks, in Proceedings of the 5th annual ACM/IEEE international conference on Mobile computing and networking, pp.445-450, 1999. [7] P. Gupta and P.R. Kumar, The Capacity Of Wireless Networks, IEEE Transection Infocom Theory, vol. 46, no. 2, pp. 388404, 2000. [8] J. Gomez, A. T. Campbell, M .Naghshineh and C. Bisdikian, Supporting Dynamic Power Controlled Routing in W i r e l e s s A d -Hoc Networks, in ACM/Kluwer Journal Wireless Network, vol.9 , pp. 443460, 2003. [9] S. Agarwal, R. H. Katz, S. V. Krishnamurthy and S. K. Dao, Distributed Power Control in Ad-Hoc Wireless Networks, in Proceeding IEEE

References

International journal Indoor and Mobile Radio Communication ,vol.2, pp.59 66, 2001. [10] M. Abolhasan, T. Wysocki and E. Dutkiewicz, A Review of Routing Protocols for Mobile Ad Hoc Networks, Ad Hoc Networks, vol. 2, pp. 1-22, 2004. [11] E. Royer, Routing in Ad-hoc Mobile Networks: On-Demand and Hierarchical Strategies, PhD thesis, University of California at Santa Barbara, California, USA, December 2000. [12] B. Liang and Z. Haas, Virtual Backbone Generation and Maintenance in Ad Hoc Network Mobility Management, in Proceeding Infocom, pp. 12931302, 2000. [13] J. Hassan and S. Jha, On the Optimization Trade-Offs of Expanding Ring Search, in Proceeding of IWDC, pp. 489494, 2004. [14] C. Perkins and E. Royer, Ad -Hoc On-Demand Distance Vector Routing, in the 2nd Annual IEEE Workshop on Mobile Computing Systems and Applications, pp. 90100, 1999. [15] C. E. Perkins and E.M. Royer, Ad Hoc On-Demand Distance Vector Routing, in Proceedings of The 2nd Annual IEEE Workshop on Mobile Computing Systems and Applications, pp. 90100 1999. [16] I. Park, J. Kim and I. Pu, Blocking Expanding Ring Search Algorithm for Efficient Energy Consumption in Mobile Ad Hoc Networks, in Proceeding WONS, pp. 185190, 2006. [17] I. Pu, Y. Shen and J. Kim, Measuring Energy-Time Efficiency of Protocol Performance in Mobile Ad Hoc Networks, in Proceeding WONS, pp. 1520, 2008. [18] J. Hassan and S. Jha, On the Optimization Trade-Offs of Expanding Ring Search, in Proceeding of IWDC, pp. 489494, 2004. [19] Y. Tseng, S. Ni, Y. Chen and J. Sheu, The Broadcast Storm Problem in A Mobile Ad Hoc Network, Springer Journal, Wireless Networks, vol. 15, no. 6, pp. 805-820, 2001. [20] J. Broch, David A. Maltz, David and B. Johnson, Yih-Chun Hu, A Performance of Multi hop Wireless Network Routing Protocol, in Mobile Computing and Networking, pp. 8597, 1998.

Dept. of CEA, GLAU, Mathura

43

References

[21]

M. Hammar and A. Pagh, Limiting Flooding Expenses in On-Demand Source-Initiated Protocols for Mobile Wireless Networks, in 18th International Parallel and Distributed Processing Symposium, 2004. S. Agarwal, R. H. Katz, S. V. Krishnamurthy a nd S. K. Dao, Distributed Routing Protocol in Ad-Hoc Wireless Networks, in Proceeding IEEE International journals, Indoor and Mobile pp.5966, Oct. 2001. Radio Communication vol.2,

[22]

[23]

K. Outsonikolas, On Optimal TTL Sequence-Based Route Discovery in Manets,in Second International Workshop on Wireless Ad Hoc Networking, pp. 78-82, 2005. S. C. Liew, Impact Of Power Control on Performance of IEEE 802.11 Wireless Networks, IEEE Transection Mobile Computing, vol. 6, no. 11, pp. 1245-1258, 2007. S. Harold and A. Vijayalakshmi, Enhanced Power Control MAC Protocol for Wireless Ad-Hoc Networks, IEEE transactions(ICCSP), vol.47, no.20, pp. 611, 2012. P. Li, H. Zhai and Y. Fang, SDMAC: Selectively Directional Routing Protocol for Wireless Mobile Ad-Hoc Networks, Springer Journal, Wireless Networks, vol. 15, no. 6, pp. 805-820, Aug. 2009 P. Minematsu and M. HOWL, An Efficient Route Discovery Scheme Using Routing History in Ad Hoc Networks, In27th Annual IEEE International Conference on Local Computer Networks, pp. 831-83, 2002. H. Zhang and Z.P. Jiang, On Reducing Broadcast Expenses in Ad Hoc Route Discovery, in Second International Workshop on Wireless Ad Hoc Networking, pp. 302-308, 2005. R. Rampant and J. Redi, A Brief Overview of Ad-Hoc Networks: Challenges and Directions, IEEE Communication Magazine, vol.40, no.5, pp.2022, 2002.

[24]

[25]

[26]

[27]

[28]

[29]

[30]

Rhee, M. Shin, S. Hong, K. Lee and S. Chong. "On the Levy Walk Nature of Human Mobility," In Proceedings of INFOCOM 2008, pp. 5-10, 2008.

[31]

M. Kim, D. Kotz and S. Kim. "Extracting A Mobility Model From Real User Traces," In Proceedings of infocom, pp. 14-21, 2006. The Network Simulator2, http://www.isi.edu/nsnam/ns, 2010.

[32]

Dept. of CEA, GLAU, Mathura

44