An Energy And Traffic Aware Routing Approach

Published Date: 02 Nov 2017

The importance of Ad-hoc network is due to its nature of infrastructure-less and decentralization. In Ad-hoc network some nodes can become a critical spot in the network as they forward the packets to most of their neighbors. These critical nodes can deplete their battery power earlier, because of excessive load and processing for data forwarding. These unbalanced loads turn to nodes failure, network partition and reduce the route lifetime and route reliability. Energy consumption issue is a vital research topic in wireless ad hoc networks, because wireless nodes in such networks operate on limited battery power. This dissertation describes An Energy and Traffic Aware Routing approach as an extension of AODV algorithm for Ad hoc network, utilizing the high energy path based on Route Selection Function. Our proposed algorithm adapts existing AODV routing protocol to improve performance in terms of traffic load on node, energy conservation and other performance metrics. The purpose of Route Selection Function is to improve lifespan of Ad hoc network and corresponding effect on overall network performance. We utilize the ability of wireless network interface cards to dynamically change their transmission power, as well as the ability of wireless devices to read the remaining battery energy and interface queue value of the device to create a table. We use interface queue as traffic on particular node if queue is less loaded than it can say that there is less traffic on node. Both these values are used to select the efficient path for data transmission. Our energy and traffic aware scheme is applied to reactive MANET routing protocols. As examples and to evaluate performance, the technique has been applied to the Ad hoc on demand Distance Vector Routing (AODV), Simulations have been carried out on mobile nodes in network. Results show improvements in network lifetime in certain mobile scenarios. Results also show better distribution of residual node energies and traffic at the end of simulations, which means that the scheme is balancing energy load and traffic more evenly across network nodes than the unmodified version of AODV.

LIST OF FIGURES

Figure No Caption Page No

2.1 Mobile Ad hoc Network 6

2.2 Classification of Routing Protocols 12

3.1 RREQ Frame Structure 34

3.2 RREQ Process of AODV 34

3.3 RREP Frame Structure 35

3.4 RREP Process of AODV 35

4.1 Initial stage of ETR-AODV 48

4.2 ETR-AODV RREQ 48

4.3 ETR-AODV RREP 49

5.1 Simple overview of ns2 50

5.2 Block diagram of ns2 architecture 52

5.3 NS Organization 53

5.4 Window of NAM 54

5.7 Simulation topology with 40 nodes 56

5.8 Status of nodes in the simulation after 244 seconds 56

5.9 End to End delay vs Mobility Speed 58

5.10 Packet Delivery Ratio vs Mobility Speed 59

5.11 Average Energy consumed vs Mobility Speed 60

5.12 Normalized Routing Load vs Mobility Speed 61

CONTENTS

Page No.

Certificate ii

Candidate’s Declaration iii

Acknowledgements iv

Abstract v

List of Figures vi

Contents vii-viii

Chapter 1. Introduction 1-4

1.1 Introduction 1

1.2 AODV Overview 2

1.3 Motivation 3

1.4 Organization of Dissertation 3

1.5 Contribution of Dissertation 4

Chapter 2. Background 5-23

2.1 Characteristics of MANET 6

2.2 MANET Applications 7

2.3 Routing 9

2.3.1 Desirable Properties of Routing Protocols 10

2.4 Classification of Ad hoc Routing Protocols 11

2.4.1 Flat Routing Protocols 12

2.4.2 Hierarchical Routing Protocols 14

2.4.3 Geographical Routing Protocols 14

2.5 Description of Some Popular Proactive Routing Protocols 15

2.5.1 Destination-sequenced distance vector (DSDV) 15

2.5.2 Optimized Link State Routing Protocol (OLSR) 16

2.6 Description of Some Popular Reactive Routing Protocols 17

2.6.1 Dynamic Source Routing Protocol (DSR) 17

2.6.2 Ad hoc On Demand Distance Vector Protocol (AODV) 18

2.7 Energy aware routing 20

2.8 Energy model 21

2.9 Energy aware metrics 22

Chapter 3. Literature Review 24-41

3.1 Introduction 24

3.2 Protocol Overview 32

3.2.1 Ad-hoc On Demand Distance Vector (AODV) 32

3.2.2 Overview 33

3.2.3 Control messages 33

3.2.4 Sequence numbers 36

3.2.5 Route discovery 36

3.2.6 Optimal TTL sequence 37

3.2.7 Link breakage 38

3.2.8 Interesting concepts of AODV 38

3.2.9 Advanced uses of AODV 39

3.2.10 Properties of AODV 39

3.3 Problem Statement 40

Chapter 4. Proposed Solution 42-49

4.1 Description of ETR-AODV 43

4.2 Pseudo-code for Proposed Algorithm 46

4.3 Explanation of ETR-AODV with example 47

Chapter 5. Simulation & Result 50-61

5.1 Simulation Environment 50

5.1.1 NS2 (NETWORK SIMULATOR 2) 50

5.1.2 Reason to choose NS2 53

5.1.3 Network Animator 53

5.2 Performance evaluation metrics 54

5.3 Simulation Topology 55

5.4 Simulation Parameters 57

5.5 Performance Evaluation 58

Chapter 6. Conclusion and Future work 62-64

6.1 Summary 62

6.2 Conclusion 62

6.3 Future work 64 REFERENCES 65-68

PUBLICATIONS 69

CHAPTER 1

INTRODUCTION

INTRODUCTION

Wireless Mobile Ad hoc network [1] enable nodes to setup a network quickly, provides advantage in deployment, cost, size and distributed intelligence over wired networks. It is a challenging task to provide the same type of services and same quality in wireless mobile environments as in wired environment. Most of earlier works on routing in ad hoc networks deal with the problem of finding and maintaining correct routes to the destination due to mobility and changing topology. Majority of routing protocols in mobile ad hoc networks use min-hop routing where the number of hops is the path length. A networks capability to provide a particular quality of service between a set of endpoints depends upon the inherent performance properties (e.g. delay, throughput, loss rate, error rate) of the links and nodes, the traffic load within the network and the control algorithms operating at different layers of the network. Ad hoc routing protocols can be broadly classified as Table driven and on demand routing protocols.

Table driven routing protocols attempt to maintain consistent, up to date routing information for the nodes in the network. In these protocols each node maintain one or more tables to store routing information, whenever any changes in network topology then they respond by propagating updates messages throughout the network in order to maintain consistent network view. On the other hand on demand routing creates routes only when required by the source node. When a node requires a route to a destination, it initiates a route discovery process within the network. The process is completed when a route is found or all possible routes have been examined. Once the route has been established, some form of route maintenance procedure maintains it until the route is no longer desired. Well-known routing protocols studied under on demand category are AODV [2] and DSR [3] (Dynamic Source Routing) while DSDV (Destination Sequenced Distance Vector routing) represents table driven category. AODV is an improvement on DSDV because it typically minimizes the number of required broadcasts by creating routes on an on-demand basis, as opposed to maintaining a complete list of routes as in DSDV routing protocol. DSR has higher connection setup delay and its performance degrades rapidly with increasing mobility as compared to AODV. AODV uses destination sequence number to find the latest route to destination requiring less setup delay also fair performance with mobility concern. Traditional routing protocols for ad hoc network select the routes under the metric of the minimum hop count. Such min-hop routing protocols can use energy unevenly among the nodes and thus it can cause some nodes to spend their whole energy earlier and make a network partition as a degradation in network performance. End to end delay and throughput are commonly used performance metrics in wired and wireless networks. Since network topology change dynamically, bandwidth and battery power are additional important factors to be considered in wireless ad hoc networks. If energy consumption is to be considered, then best suited protocols are on-demand protocols where control information transfer is limited and is not as frequently updated as in table driven protocols or proactive protocols. On demand protocols flooding the route request packets throughout the network does the route discovery. Our main aim is to make the routing protocol more energy efficient and ultimately to perform traffic aware of the node in the ad hoc network.

MOTIVATION

The traditional routing protocol did not consider the energy consumption of nodes and traffic load on nodes, so there need to design an energy and traffic aware algorithm to improve network performance. The rational of our inspiration is that most of the enhancement in AODV protocol has been based on shortest path or multipath. There has been several works in on demand routing protocols which considered some factor in unified way. This motivates us to develop a new energy and traffic aware Ad-hoc On Demand Distance Vector Routing protocol, to make an efficient route selection on the basis of remaining energy and traffic load of each node. Therefore, we propose routing mechanism i.e. "An Energy and Traffic Aware Routing Approach as an Extension of AODV". After in depth analysis of literature, it is vivid that due to varying nature of MANET, no single routing protocol works efficiently in all conditions of networks.

1.4 ORGANIZATION OF DISSERTATION

Rest of the dissertation is organized as follows. Chapter 2 presents Background, in which a broad classification of the existing routing protocols in mobile ad-hoc network and energy based routing as well as traffic load balancing have been discussed. Chapter 3 presents the Literature review and formation of problem. In chapter 4 we present our proposed approach in which first we give description of proposed approach and explain it with an example. Chapter 5 includes simulation methodology and analysis of results. Finally chapter 6 provides Conclusion and future extensions.

1.5 CONTRIBUTION OF DISSERTATION

This dissertation proposes a solution which is considering the power level of node as well as congestion on node. So, our algorithm performs better for finding out the routing path with better energy and low congestion path. Our algorithm tries to establish balance of energy among the nodes as well as traffic load in the path. The proposed ETR-AODV achieve enhancement over traditional AODV as listed below:

Enhance route selection method on the basis of residual energy and traffic load of each node.

We define power factor as the ratio of total available energy of nodes in path and total initial energy of all the nodes in that path.

Also define congestion factor as percentage of the network interface queue that is vacant.

Chapter 2

BACKGROUND

This chapter presents Mobile ad hoc network, their characteristics and their applications. Routing and their limitations in ad hoc network are also discussed in this chapter. After that we discussed different routing protocols, their merits and demerits. Their common comparison table shows their characteristics with different parameters. We also discussed energy aware routing and energy model used in ad hoc networks.

A wireless Mobile ad-hoc network is a collection of mobile nodes with no pre-established infrastructure, forming a temporary network without the aid of any centralized administration, in which each node cooperates by forwarding packets to each other to allow nodes to communicate beyond direct wireless transmission range. Participating nodes have a wireless interface and communicates with each other over either radio or infrared. Laptop, computers and personal digital assistants (PDA) that communicates directly with each other are some examples of nodes in an ad-hoc network. Nodes in the ad-hoc network are normally mobile, but can also consist of stationary nodes as access points to the Internet. Semi mobile nodes can be used to deploy relay points in areas where relay points might be needed temporarily.

Figure 2.1 shows a simple ad-hoc network with three nodes. The node 3 is not within transmitter range of node 2. However the middle node 5 or 6 can be used to forward packets between the nodes. These middle nodes are acting as a router and the four nodes have formed an ad-hoc network.

This is to be sure that the network will not collapse just because one of the mobile nodes moves out of transmitter range of the others. Nodes are able to enter/leave the network as their movable characteristics. Because of the limited transmitter range of the nodes, multiple hops may be needed to reach other nodes. Every node willing to participate in an ad-hoc network must be willing to forward packets for other nodes. Thus every node acts both as a host and as a router. A node can be viewed as an abstract entity consisting of a router and a set of mobile hosts (Figure 2.1). A router is an entity, which, among other things runs a routing protocol.

3

6

1

4

2

5http://monet.postech.ac.kr/images/introduction/image005.jpg

Fig 2.1: Mobile Ad hoc Network

We can formalize the above statement by defining an ad hoc network as an autonomous system of mobile nodes (serving as routers) connected by wireless links, which forms a communication network in the form of an arbitrary graph with mobile nodes. These nodes may be located in or on airplanes, ships, trucks, cars, perhaps even on people or very small devices. This is in contrast to the well-known single hop cellular network model that supports the needs of wireless communication by installing base stations as access points. In these cellular networks, communications between two mobile nodes completely rely on the fixed base stations and the wired backbone. In a MANET, no such infrastructure exists due to dynamic network topology formation in an unpredictable manner since nodes are free to move.

2.1 CHARACTERISTICS OF MANET

MANETs have several salient characteristics:

1) Dynamic topologies: Nodes are free to move arbitrarily; thus, the network topology which is typically multi-hop may change randomly and rapidly at unpredictable times, and may consist of both bidirectional and unidirectional links.

2) Bandwidth-constrained, variable capacity links: Wireless links will continue to have significantly lower capacity than their hardwired counterparts. In addition, the realized throughput of wireless communications after accounting for the effects of multiple access, fading, noise, and interference conditions ,etc. it is often much less than a radio's maximum transmission rate. One effect of the relatively low to moderate link capacities is that congestion is typically the norm rather than the exception, i.e. aggregate application demand will likely approach or exceed network capacity frequently.

3) Energy-constrained operation: Some or all of the nodes in a MANET may rely on batteries or other exhaustible means for their energy. For these nodes, the most important system design criteria for optimization may be energy conservation.

4) Limited physical security: Mobile wireless networks are generally more prone to physical security threats than are fixed- cable nets. The increased possibility of eavesdropping, spoofing, and denial-of-service attacks should be carefully considered. Existing link security techniques are often applied within wireless networks to reduce security threats.

2.2 MANET APPLICATIONS

An ad hoc application is a self-organizing application consisting of mobile devices forming a peer-to-peer network where communications are possible because of proximity of the devices within a physical distance. MANET can be used to form the basic infrastructure for ad hoc applications. There are many applications to ad hoc networks. As a matter of fact, any day-to-day application such as electronic email and file transfer can be considered to be easily deployable within an ad hoc network environment. Web services are also possible in case any node in the network can serve as a gateway to the outside world. In this discussion, we need not emphasize the wide range of military applications possible with ad hoc networks. Not to mention, the technology was initially developed keeping in mind the military applications, such as battlefield in an unknown territory where an infrastructure network is almost impossible to have or maintain. In such situations, the ad hoc networks having self-organizing capability can be effectively used where other technologies either fail or cannot be deployed effectively. Advanced features of wireless mobile systems, including data rates compatible with multimedia applications, global roaming capability, and coordination with other network structures, are enabling new applications

2.3 ROUTING

As mobile ad hoc networks are characterized by a multi-hop network topology that can change frequently due to mobility, efficient routing protocols are needed to establish communication paths between nodes, without causing excessive control traffic overhead or computational burden on the power constrained devices [6]. A large number of solutions have already been proposed, some of them being subject to standardization within the IETF. A number of proposed solutions attempt to have an up-to-date route to all other nodes at all times. To this end, these protocols exchange routing control information periodically and on topological changes. These protocols, which are called proactive routing protocols, are typically modified versions of traditional link state or distance vector routing protocols encountered in wired networks, adapted to the specific requirements of the dynamic mobile ad hoc network environment. Most of the time, it is not necessary to have an up-to-date route to all other nodes. Therefore, reactive routing protocols only set up routes to nodes they communicate with and these routes are kept alive as long as they are needed. Combinations of proactive and reactive protocols, where nearby routes (for example, maximum two hops) are kept up-to-date proactively, while far-away routes are set up reactively, are also possible and fall in the category of hybrid routing protocols. A completely different approach is taken by the location-based routing protocols, where packet forwarding is based on the location of a node’s communication partner. Location information services provide nodes with the location of the others, so packets can be forwarded in the direction of the destination.

2.3.1 Desirable Properties of Routing Protocols

If the conventional routing protocols do not meet our demands, we need a new routing protocol. The question is what properties such protocols should have? These are some of the properties [5] that are desirable:

Distributed operation

The protocol should of course be distributed. It should not be dependent on a centralized controlling node. This is the case even for stationary networks. The difference is that nodes in an ad-hoc network can enter/leave the network very easily and because of mobility the network can be partitioned.

Loop free

To improve the overall performance, we want the routing protocol to guarantee that the routes supplied are loop-free. This avoids any waste of bandwidth or CPU consumption.

Demand based operation

To minimize the control overhead in the network and thus not wasting network resources more than necessary, the protocol should be reactive. This means that the protocol should only react when needed and that the protocol should not periodically broadcast control information.

Unidirectional link support

The radio environment can cause the formation of unidirectional links. Utilization of these links and not only the bi-directional links improves the routing protocol performance.

Security

The radio environment is especially vulnerable to impersonation attacks, so to ensure the wanted behavior from the routing protocol, we need some sort of preventive security measures. Authentication and encryption is probably the way to go and the problem here lies within distributing keys among the nodes in the ad-hoc network. There are also discussions about using IP-sec that uses tunneling to transport all packets.

Power conservation

The nodes in an ad-hoc network can be laptops and thin clients, such as PDAs that are very limited in battery power and therefore uses some sort of stand-by mode to save power. It is therefore important that the routing protocol has support for these sleep-modes.

Multiple routes

To reduce the number of reactions to topological changes and congestion multiple routes could be used. If one route has become invalid, it is possible that another stored route could still be valid and thus saving the routing protocol from initiating another route discovery procedure.

Quality of service support

Some sort of Quality of Service support is probably necessary to incorporate into the routing protocol. This has a lot to do with what these networks will be used for. It could for instance be real-time traffic support.

None of the proposed protocols from MANET have all these properties, but it is necessary to remember that the protocols are still under development and are probably extended with more functionality. The primary function is still to find a route to the destination, not to find the best/optimal/shortest-path route.

2.4 Classification of Ad hoc Routing Protocols

Routing protocol in MANET can be classified into several ways depending upon their network structure, communication model, routing strategy, and state information and so on but most of these are done depending on routing strategy and network structure [3,10].

Fig 2.2: Classification of Routing Protocols

Based on the routing strategy the routing protocols can be classified into two parts: 1.Table driven and 2. Source initiated (on demand) while depending on the network structure these are classified as flat routing, hierarchical routing and geographic position assisted routing [3]. Flat routing covers both routing protocols based on routing strategy. The classification of routing protocols is shown in the Figure.

2.4.1 Flat Routing Protocols

Flat routing protocols distribute information as needed to any router that can be reached or receive information. No effort is made to organize the network or its traffic, only to discover the best route hop by hop to a destination by any path. Think of this as all routers sitting on a flat geometric plane. These protocols are mainly divided into two classes, First one is proactive routing that is also known as table driven and other is reactive that’s called on-demand routing protocols. Common thing is general for both protocol classes is that every node participates in routing play an equal role. These are further been classified based on their design principles; Proactive routing is mostly based on LS (Link State) while on-demand routing based on DV (Distance Vector).

Pro-Active /Table Driven routing Protocols

Proactive protocols continuously learn the topology of the network by exchanging topological information among the network nodes. Thus, when there is a need for a route to a destination, such route information is available immediately. If the network topology changes too frequently, the cost of maintaining the network might be very high. If the network activity is low, the information about actual topology might even not be used. Proactive protocols continuously evaluates the routes within the network so that when we are required to forward the packet route is already known and immediately ready for use. So there is no time delay. So a shortest path can be find without any time delay however these protocols are not suitable for very dense ad-hoc networks because in that condition problem of high traffic may arise. Several modifications of proactive protocols have been proposed for removing its shortcomings and use in ad-hoc networks. It maintains the unicast routes between all pair of nodes without considering of whether all routes are actually used or not.

Examples of Proactive MANET Protocols include Optimized Link State Routing (OLSR), Fish-eye State Routing (FSR), Destination-Sequence Distance Vector (DSDV),Cluster-head Gateway Switch Routing Protocol (CGSR) etc.

Reactive Routing Protocols

The reactive routing protocols are based on some sort of query-reply dialog. It is also called on demand routing. It is more efficient than proactive routing and most of the current work and modifications have been done in this type of routing for making it more and more better. The main idea behind this type of routing is to find a route between a source and destination whenever that route is needed whereas in proactive protocols we were maintaining all routes without regarding its state of use. So in reactive protocols we don’t need to bother about the routes which are not being used currently. This type of routing is on demand. Discovering the route on demand avoids the cost of maintaining routes that are not being used and also controls the traffic of the network because it doesn’t send excessive control messages which significantly create a large difference between proactive and reactive protocols. Time delay in reactive protocols is greater comparative to proactive types since routes are calculated when it is required.

Examples of Proactive MANET Protocols include Ad hoc On Demand Distance Vector (AODV), Dynamic Source Routing Protocol (DSR), Temporally Ordered Routing Algorithm (TORA), Dynamic MANET On-Demand (DYMO), Associativity Based Routing (ABR), Signal Stability-Based Adaptive Routing (SSA), Location-Aided Routing Protocol (LAR) etc.

Hybrid Routing Protocols

Both of the proactive and reactive routing methods have some pros and cons. In hybrid routing a well combination of proactive and reactive routing methods are used which are better than the both used in isolation. It includes the advantages of both protocols. As an example facilitate the reactive routing protocol such as AODV with some proactive features by refreshing routes of active destinations which would definitely reduce the delay and overhead so refresh interval can improve the performance of the network and node. So these types of protocols can incorporate the facility of other protocols without compromising with its own advantages. Examples of hybrid protocols are Zone Routing Protocol (ZRP).

Examples of Hybrid Protocols are Zone Routing Protocol (ZRP), Wireless Ad hoc Routing Protocol (WARP) etc.

2.4.2 Hierarchical Routing Protocols

A hierarchical control structure is employed by the protocols which fall in this class [7]. The nodes located in a common scope (may be defined by their distances to each other) in the network are grouped together into a cluster, so that the network is defined as clusters. The nodes of a specific cluster elect a cluster head who coordinates the work between the different nodes in the cluster. This clustering can be extended to a multi level hierarchy. A very important example of this class is Cluster-Head Gateway Switch Routing Protocol (CGSR). A hierarchical addressing scheme is required [1]. Some Hierarchical Routings are Hierarchical State Routing (HSR), Zone Routing Protocol (ZRP), Cluster-head Gateway Switch Routing Protocol (CGSR), Landmark Ad hoc Routing Protocol (LANMAR) etc.

2.4.3 Geographical Routing Protocols

Geographic routing protocols scale better for ad hoc networks mainly for two reasons: 1.) There is no necessity to keep routing tables up-to-date and 2.) No need to have a global view of the network topology and its changes. Therefore, geographic routing protocols have attracted a lot of attention in the field of routing protocols for MANETs. These geographic approaches allow routers to be nearly stateless because forwarding decisions are based on location information of the destination and the location information of all one-hop neighbours. Most of these protocols keep state only about the local topology (i.e., neighbor’s location information). No routing table

is constructed. As a result, establishment and maintenance of routes are not required, reducing the overhead considerably.

Some Geographical Routing Protocols are Geographic Addressing and Routing (GeoCast) , Distance Routing Effect Algorithm for Mobility (DREAM) and Greedy Perimeter Stateless Routing (GPSR) etc.

2.6 Description of Some Popular Reactive Routing Protocols:

2.6.2 AODV (Ad hoc On Demand Distance Vector Protocol):

Ad hoc On Demand Distance Vector AODV [9] is a variation of Destination-Sequenced Distance-Vector (DSDV) routing protocol which is collectively based on DSDV and DSR. It aims to minimize the requirement of system-wide broadcasts to the greater extent. It does not maintain routes from every node to every other node in the network rather they are discovered as and when needed and are maintained only as long as they are required. The key steps used by AODV for establishment of unicast routes are Route discovery and Route maintenance.

i) Route Discovery

When a node wants to send a data packet to a destination node, the entries in route table are checked to ensure whether there is a current route to that destination node or not. If it is there, the data packet is forwarded to the appropriate next hop toward the destination. If it is not there, the route discovery process is initiated. AODV initiates a route discovery process using Route Request (RREQ) and Route Reply (RREP). The source node will create a RREQ packet containing its IP address, its current sequence number, the destination’s IP address, the destination’s last sequence number and broadcast ID. The broadcast ID is incremented each time the source node initiates RREQ. Basically, the sequence numbers are used to determine the timeliness of each data packet and the broadcast ID & the IP address together form a unique identifier for RREQ so as to uniquely identify each request. The requests are sent using RREQ message and the information in connection with creation of a route is sent back in RREP message. The source node broadcasts the RREQ packet to its neighbors and then sets a timer to wait for a reply. To process the RREQ, the node sets up a reverse route entry for the source node in its route table. This helps to know how to forward a RREP to the source. Basically a lifetime is associated with the reverse route entry and if this entry is not used within this lifetime, the route information is deleted. If the RREQ is lost during transmission, the source node is allowed to broadcast again using route discovery mechanism.

ii) Route maintenance

As long as the route remains active, it will continue to be maintained. A route is considered active as long as there are data packets periodically travelling from the source to the destination along that path. Once the source stops sending data packets, the links will time out and eventually be deleted from the intermediate node routing tables. If a link break occurs while the route is active, the node upstream of the break propagates a route error (RERR) message to the source node to inform it of the now unreachable destination(s). After receiving the RERR, if the source node still desires the route, it can reinitiate route discovery.

Advantages of AODV

The benefits of AODV protocol are that it favors the least congested route instead of the shortest route and it also supports both unicast and multicast packet transmissions even for nodes in constant movement.

It also responds very quickly to the topological changes that affects the active routes.

AODV does not put any additional overheads on data packets as it does not make use of source routing.

Limitations of AODV

The limitation of AODV protocol is that it expects/requires that the nodes in the broadcast medium can detect each others’ broadcasts. It is also possible that a valid route is expired and the determination of a reasonable expiry time is difficult. The reason behind this is that the nodes are mobile and their sending rates may differ widely and can change dynamically from node to node. In addition, as the size of network grows, various performance metrics begin decreasing.

AODV is vulnerable to various kinds of attacks as it based on the assumption that all nodes must cooperate and without their cooperation no route can be established.

Ad hoc routing protocols can broadly be classified as Table driven and on demand protocols [12, 13]. Table driven routing protocols attempt to maintain consistent, up to date routing information from each node to every other node in the network. These protocols require each node to maintain one or more tables to store routing information, and they respond to changes in network topology by propagating updates throughout the network in order to maintain consistent network view. On demand routing creates routes only when desired by the source node. When a node requires a route to a destination, it initiates a route discovery process within the network. The process is completed once a route is found or all possible route permutations have been examined. Once the route has been established, some form of route maintenance procedure maintains it until the route is no longer desired. Broad classification of above two types of protocols is highlighted in Table 1. Prominent routing protocols studied under on demand category are AODV and Dynamic Source Routing (DSR) while Destination Sequenced Distance Vector routing (DSDV) represents table driven category. Studying prominent features, it is found that AODV is an improvement on DSDV because it typically minimizes the number of required broadcasts by creating routes on an on-demand basis, as opposed to maintaining a complete list of routes as in DSDV. DSR has higher connection setup delay and its performance degrades rapidly with increasing mobility as compared to AODV. AODV uses destination sequence number to find the latest route to destination requiring less setup delay also fair performance with mobility issue.

Parameters

On-demand

Table driven

Availability of routing information

Available when needed

Always available

Routing philosophy

Flat

Mostly Flat

Periodic route updates

Not required

Essential

Coping with mobility

Using localized route discovery

Inform other nodes to achieve consistent routing table

Signaling traffic generated

Grows with increasing mobility of active routes

Greater than that of on-demand routing

Quality of service support

Few can support QoS

Mainly shortest path as QoS metric

Table 2.2: Comparison between On-demand vs Table driven

2.7 Energy aware routing:

The aim of energy-aware routing protocols is to reduce energy consumption in transmission of packets between source and a destination, to avoid routing of packets through nodes with low residual energy, to optimize flooding of routing information over the network and to avoid interference and medium collisions. A single node failure in sensor networks is usually unimportant because it does not lead to a loss of sensing and communication coverage whereas ad-hoc networks are oriented towards personal communication and the loss of connectivity to any node is significant. There have been increased interests in routing algorithms that take the energy levels of the nodes into account. Nodes in wireless ad hoc networks typically use batteries as their power source. These batteries often have a limited capacity, which is depleted with every transmission. It is important to consider this when designing routing protocols for ad hoc networks. Only finding the paths that require the least amount of power is not enough; some nodes might still be used much more than others, and will therefore be prematurely depleted.

Many energy efficient routing protocol proposals were originally studied for sensor networks [11], where the limited energy of nodes is a strong constraint; in MANET, however, the requirements are different: a node has generally more hardware resources (capable of better performance, but consuming more energy) and the protocol must preserve the resources of every node in the network (not only a subset of them, because each node can be, at any time, source or destination of data). A single node failure in sensor networks is usually unimportant if it does not lead to a loss of sensing and communication coverage; ad-hoc networks, instead, are oriented towards personal communication and the loss of connectivity to any node is significant. In the routing protocol design of mobile nodes, many issues need to be considered in order to offer many important properties such as scalability, QoS support, security, low power consumption and so on. In this chapter we focus on the energy issues facing some important aspects going from the energy model definition for the computation of the energy consumption to energy-aware metrics definition and routing protocol design. If a network composed of mobile nodes communicating using a wireless radio and where each node can communicate with each other using the other mobile nodes as relay nodes is applied in a communication system, many challenging design issues need to be addressed. MANET technology became, in the last years, more commercial in comparison with the past where it was used for military purpose and this implies more additional features to offer to the end user with particular reference to quality of service, security and to node lifetime duration. In this chapter energy saving techniques at network layer and the routing strategies that allow better energy expenditure and load distribution in order to prolong the network lifetime are considered.

2.8 Energy model:

A wireless network interface can be in one of the following four states: Transmit Receive, Idle or Sleep. Each state represents a different level of energy consumption.

Transmit: A node is transmitting a frame with some transmission power.

Receive: A node is receiving a frame with some reception power. That energy is consumed even if the frame is discarded by the node because it was intended for another destination, or it was not correctly decoded.

Idle (listening): Even when no messages are being transmitted over the medium, the nodes stay idle and keep listening the medium.

Sleep: when the radio is turned off and the node is not capable of detecting signals, no communication is possible. The node uses the power that is largely smaller than any other power

2.9 Energy aware metrics

The majority of energy efficient routing protocols for MANET try to reduce energy consumption by means of energy efficient routing metric, used in routing table computation instead of the minimum-hop metric there are four possibilities to save power from the devices:

Minimal Energy Consumption per Packet

The energy consumption is the sum of power consumed on every hop in the path from a packet. The power consumption on a hop is a function of the distance between the neighbor and the load of this hop. So it is interesting to choose a route where the distance between the nodes isn't too long and also it is interesting to take a shorter route so there aren't too many hopes on the route where the power level gets down.

Maximize Network Connectivity

This metric tries to balance the load on all the nodes in the network. This assumes significance in environment where the network connectivity is to be ensured.

Minimum Variance in Node Power Levels

This metric proposes to distribute the load among all nodes so that the power consumption remains uniform to all nodes. This problem is very complex when the rate and size of data packets vary. When every node has the same level in power, you can be sure that the network functions longer. Because when there is a node which has to switch

off because of the power level the whole network is in danger and it can break down the connectivity between the nodes.

Minimize Maximum Node Cost

This metric minimizes the maximum cost per nodes for a packet after routing a number of packets or after a specific period. So a node can be blocked for routing to save battery power. This metrics saves the connectivity from every node. When a node has been used several times for route, it blocks itself to save the power.

CHAPTER 3

LITRATURE REVIEW

This chapter includes the literature what the other author’s point of view towards the energy and traffic load in their papers. This short literature will show the view of different author’s problem as well as their solutions.

3.1 INTRODUCTION

As in Ad hoc network the nodes can make a network without any central access point, so there need a different type of routing protocols apart from the conventional routing protocols as in wired networks. Most of previous work on routing in wireless ad hoc networks deals with the problem of finding and maintaining correct routes to the destination during mobility and changing topology. Majority of routing protocols in mobile ad hoc networks use shortest path length routing [32] where the number of hops is the path length. A networks ability to provide a specified quality of service between a set of endpoints depends upon the inherent performance properties (e.g. delay, throughput, loss rate, error rate) of the links and nodes, the traffic load within the network and the control algorithms operating at different layers of the network. Conventional routing protocols for ad hoc networks select the routes under the metric of the minimum hop count. Such min-hop routing protocols can use energy unevenly among the nodes and thus it can cause some nodes to spend their whole energy earlier. Mobile ad hoc networks have additional issues like dynamic topology, asymmetric links, routing overhead and interference due to nearby transmission. Considering issue of mobility of nodes, it is desirable to have route updates in order to know the current status of the node position in the network. This feature demands table driven protocols. However if energy consumption is to be considered, one has to go for on demand protocol where control information transfer is limited and is not as frequently updated as in table driven protocols. In on demand protocols, flooding the route request packets throughout the network does the route discovery. Our approach is to save energy consumption at the cost of reduced route updates. The approach is based on the reduction of energy consumption during the connection request phase, passing by phase of route discovery, as well as during the phase of data transfer and finishing by the phase of route maintenance. Our motto is showing better results for Number of alive nodes, its capacity to balance the consumption of energy on the totality of the network.

Many routing protocols exist in ad hoc Network, these protocols are based on min-hop count but it may not be the best routing criteria with respect to the energy and traffic load of a node because wireless nodes in Ad hoc networks operate on limited battery power. AODV [14] and DSR [15] are min hop routing protocols. Tai Hieng [16] proposed an algorithm which is an improvement of energy efficient routing by selecting high energy paths, taking account of energy conservation and other performance metrics. Energy metric is used as route selection method to improve lifespan of ad hoc network. Route selection is one of the main domains where many researchers are trying to improve performance of Ad hoc network. In [17] the authors proposed a

new metric which consider as the stability of the paths by taking three parameters viz "affinity", "available bandwidth" and "battery level" in routing decisions. It also maintains multipath to achieve load sharing. The routing table in AODV, maintains only one route to specified node, therefore the source node needs to reinitiate route discovery process as a route fails. Luo Chao & Lipingan [21] presented an improved method, each source will maintain backup routes, when the primary route fails the source node will use the backup route to send packets, which improves the packet delivery ratio and end to end delay. According to different route selection mechanism, energy based routing algorithm can be categories as, minimum total energy consumption which select the route with minimum total energy consumption between source to destination, In which some nodes taking part in transmission may run out of power and leads the partition of networks. Another is maximizing network lifetime algorithm, it establish the route by avoiding the lower energy node so that it can balance the energy consumption of nodes and enhance the lifetime of networks. The final is mixed optimizing routing algorithm which mix the above two categories. Yonghui Chen [18] proposed energy saving routing protocol named EEAODV based on AODV. In routing discovery process source node consider the remaining energy of intermediate nodes and also consider the influence of these nodes, which may cause route changed. In [19] the authors have given an effective scheme to balance the node in network. This new scheme can be

applied in most On-demand routing protocols, It is implement in the process of route request, when RREQs packets are flooded to establish routes, only the qualified nodes, which have a potential to serve as intermediate forwarding nodes will respond to these packets. So that the established path will not be very congested and traffic load will be distributed evenly in the network it also considers a threshold value, which is used to judge whether intermediate node is overloaded, is variable and changing along the nodes interface queue occupancy around the backward path. Most of ad-hoc routing protocols do not consider contention time, occurs in medium reservation procedure. Long contention times can be more critical than hop counts in determining the end to end delay. Some mobile nodes may lead to long queuing delays, low packet delivery ratio and inefficient power consumption. Bong Chan Kim [20] proposed routing protocol with minimum contention time and load balancing (MCL). This protocol has two main functions, MCL selects a route with minimum contention among many possible routes in the route selection procedure and secondly intermediate nodes do not reply to RREQs in the route discovery procedure. MCL outperforms in term of packet delivery ratio, average end to end delay, and normalized routing overhead. Unbalanced traffic may lead to more delay, packet dropping and decreasing packet delivery ratio and unbalanced energy consumption leads to node

failure, network partitioning, and decrease route reliability. Author in [22] suggested an energy efficient Ad-hoc On-demand routing (EEAODR) algorithm that balances energy load so that a minimum energy level is maintained among nodes and the network lifetime is increased by distributing energy consumption in the network.

Battery power is a limited and precious resource in MANET, and it is expected that battery technology is not likely to growing as fast as computing and communication technologies do. Hence, extend the lifetime of batteries is an important issue, especially for MANET, which is totally supported by batteries. Based on this observation Jin-Man Kim [23] proposed enhanced AODV (Ad-hoc On-demand Distance Vector) routing protocol which is modified to improve the networks lifetime in MANET (Mobile Ad-hoc Network).They make a improvement for the AODV protocol is to maximize the networks lifetime by applying an Energy Mean Value algorithm which caring node energy-aware. Also amplify the entire network lifetime through the delaying method of RREQ flooding by considering the node's energy state & the entire node's Energy Mean Value. They attempted to extend the entire network lifetime by adjusting RREQ delay time according to the data acquired from comparison between node’s energy states and the entire network's Energy Mean Value.

Yang Qin et.al[24] propose a routing scheme which considered power conservation, shortest path and traffic load balancing, named power and traffic balance awareness paths selection routing scheme ( PTPSR ). In this routing scheme, they considered both the shortest path and the power conservation in a unified way. They define an energy factor as the ratio of the remaining energy over the initial energy of a node. They use the products of the energy factors of all the nodes along different paths as the selection criteria. They compare the PTPSR with other routing protocols, e.g., ad hoc on demand multi-path distance vector (AOMDV). In future work, they will try to incorporate other energy conservation techniques, such as switching nodes with low energy levels to listening mode where nodes are able to receive data but not transmit, into their protocol. This will be helpful to get maximized lifetime for the network.

In [25] the authors focus on the performance study of four routing protocols, namely AODV, AODVUU, RAODV and AOMDV. These protocols are from AODV family, as all these protocols consider AODV as the base routing protocol upon which these protocols are improved. They investigated whether a multiple path algorithm like AOMDV would result in more data delivery as compared to single path solutions like AODV in a sensor network. Also, checked the reverse route discovery mechanisms engaged in RAODV for a sensor network. They also compare the performance of all these four routing protocols, namely AODV, AODVUU, AOMDV and RAODV under various network scenarios. These protocols from a sensor network point of view by widely using various performance metrics like packet delivery ratio, average network delay, network throughput and normalized routing load. They reveal that AOMDV and RAODV show good performance when compared to AODV in an ad hoc network environment, same cannot be said when the routing protocols are applied for a sensor network. A finer design does not guarantee a big boost in the performance in a different environment as they discuss in their paper. They also suggest that instead of using the default AODV routing protocol that comes with NS-2 for simulation purpose to use the AODVUU implementation for a ZigBee/802.15.4 standard scenario. Their future work includes designing a new routing protocol that takes in to consideration the various challenges under which a routing protocol has to work in a unique and challenging sensor environment.

In [26] they improved the Dynamic Source Routing (DSR) protocol to Load Balanced DSR (LBDSR) protocol. In this paper modified control messages in DSR in order to maintain remained energy of intermediate nodes. Routes discovered by these modified control messages, have a remained energy array field which demonstrates total of remained battery power of all nodes in a route. Selecting a route is depending on length, freshness, traffic and energy level of routes. Author’s protocol shows better traffic balancing, energy consumptions balancing, and end-to-end delay metrics than DSR and can be customized to achieve even better performance related to a specific metric. LBDSR modified control packets, route tables, and route selection method in DSR protocol to achieves a higher performance in load balancing. The proposed approach can apply to many of the current routing protocols especially, on the other important reactive protocol namely, AODV.

In this paper [27], Authors proposed two methods for improve the AODV protocol. A new multipath routing protocol that uses all discovered path in chorus for transmitting data, by using this approach data packets are balanced over discovered paths and energy consumption is distributed across many nodes through network. Also authors proposed a stability evaluation method and applied that in an optimized version of ad hoc on demand distance vector (AODV) routing algorithm by doing some modification at RAODV algorithm. They proposed an algorithm namely Modified Reverse Ad Hoc On-demand Vector (MRAODV), the route request packet have no change and it is same as AODV, but rout reply packet changed for route stability estimation purpose. Authors applied link stability in RAODV for decrease overhead of discovery and maintenance of routing, and increased the packet delivery ratio in mobile ad hoc networks. Authors introduced a new multipath routing that discovered paths simultaneously; this technique applied to AODV and evaluated via several simulations scenarios. Results show that our protocol have better packet delivery ratio in compare of AODV and AOMDV, also the energy consumption is distributed across many nodes that cases the network life time in LBAODV is better than AODV and AOMDV. In future they are going to observe their work over other routing protocols such as DSR and evaluate its performance with different scenarios. Also this paper presented a new protocol for mobile ad hoc networks based on link stability and reverse packet transmission. Authors changed RAODV routing algorithm and made an optimized version of AODV. New method shows good performance in some ways. In MRAODV they changed route replay packet configuration of RAODV and named it RRREQ. These packets should be transmitted to destination node for building multiple routes.

Wireless Sensor Networks (WSNs) have inherent and distinctive characteristics. One of the most important concerns is their energy constraint. Energy aware routing protocol is very vital in WSN [28], but routing protocol which simply considers energy has no efficient performance. Congestion can affect routing protocol performance. Congestion occurrence in network nodes leads to increasing packet loss and energy consumption. Another parameter which affects routing protocol efficiency is fairness in nodes energy consumption. When fairness is not considered in routing process, network will be partitioned very soon due to energy drainage and then the network performance will be decreased. Authors proposed a Hierarchical Tree based Energy efficient and Congestion aware Routing Protocol (HTECRP). The proposed protocol is an energy efficient routing in which they try to manage congestion and perform fairness in network.

For congestion management they presented a new hierarchical energy efficient routing protocol for sensor networks. In this effort Routing protocol divides network into many clusters, then using Dijkstra algorithm constructs a routing tree for each cluster. In routing tree, most number of children for cluster nodes is determined. Proposed protocol using routing tree and node’s neighbors average queue length as a parameter manages congestion. The effectiveness of the protocol is validated by simulation. Proposed protocol considers only intra cluster routing; for future work they are extending the protocol to perform routing inter clusters.

In [29] Kordafshari proposed an approach for routing in SPEED protocol considering residual energy in routing decisions. Due to the limited energy of a sensor node, energy efficient routing is a very significant issue in sensor networks. This approach finds energy-efficient paths for delay constrained data in real-time traffic. The SPEED protocol does not consider any energy metric in its routing. In their approach, routing is based on a weight function, which is a combination of the three factors: Delay, Energy & Speed. Here, the node with the greatest value in the weight function is to be selected as the next hop forwarding. Protocol focus on increase the network lifetime by considering energy metric in routing decisions. This method aims to construct a nearly stateless routing protocol, which can be used to route data based on the nodes’ residual energy. Simulation results and a comparison of this algorithm with the SPEED protocol represent that distributing energy consumption on nodes in routing will protect the nodes with less energy and prevents from a fast destruction. This, directly, causes the network lifetime to increase. However, a problem should be considered in the SPEED protocol, furthermore: How can we guarantee the successful delivery of the data packets in such a randomly distributed sensor network? For instance, in an extreme case where the sink node happens to be a set in an isolated place, how can we deliver the data packets?

In ad hoc networks some nodes can became a critical spot in the network because they support packet forwarding for most of their neighbours. Critical nodes would consume more energy due to the extra load and exhaust battery sooner. These unfairly loaded nodes will lead to node failure, network partitioning, decrease in route lifetime and route reliability. To avoid this problem, [30] proposes a new routing protocol, called Energy Efficient DSR (E2DSR) for balancing the energy consumption amongst the nodes in the network. E2DSR uses some mechanisms of Dynamic Source Routing (DSR) but defines a new formation for control packets, changes the routing behavior in nodes, implements a new "Energy Table" and creates a whole new algorithm for route cache and selection. This new protocol is capable of balancing power consumption amongst different nodes in the network effectively delaying earlier node failure due to battery exhaustion and increasing route reliability. E2DSR uses small routing tables and formulas for route priority computation and a simple on demand discovery mechanism. E2DSR was conceived to be a lightweight protocol so it can also be used in sensor networks. Future work in E2DSR includes a full evaluation of protocol performance, using the hereby described metrics, and protocol scalability, by implementing it in a larger scenario. E2DSR is continuously being fine tuned according to the results received by both our simulation platform and implementation, currently the energy field in E2DSR uses a linear quantization, however latest studies indicate that if we give more granularity to the lower levels of battery energy, by using a non-linear quantization method, we will be able to achieve a more effective energy balance.

Most improvements are made by modifying route discovery mechanism by taking into account different network parameters like; Node status, power consumption, link status, packet overhead, congestion and others. In [31] the protocol selects route on the basis of traffic load on the node and resets path as the topology changes. Instead of transmitting entire data through one route, new efficient paths are discovered from time to time during transmission. This is an efficient technique for transmissions that requires a link for longer period of time. For the calculation of load on each node during the process of route discovery, this protocol modified the basic AODV route discovery mechanism. When a source node wants to communicate with another node for which it has no routing information in its routing table then route discovery process starts. Source node broadcasts RREQ message to its neighbors. When a neighbor receive RREQ message it will calculate the number of packets in the queue and divide it with the size of the queue and add the value in the reserved field of the RREQ message. This process is done at each node in the route to the destination. At the destination the average Load ratio is calculated by dividing the reserved field value with the number of hop count. The destination decides on the basis of average value that to which route it has to send reply. Protocol reveals enhanced route discovery mechanism that ensures shortest routing path with relative to time, so the source sends packets quickly to destination then basic AODV. As [31] showed sharing of load decreases the network congestion which directly leads to the decrease of overflowing of queuing buffer and packets loss. So packet delivery ratio and throughput is increased. Proposed protocol is efficient for a transmission that requires a link for longer period of time.

A major bottleneck in Mobile Ad hoc networks (MANETs) is the energy consumption since nodes are usually mobile and battery operated. To maximize the lifetime of mobile ad