A Infrastructure Less Networks Has Found Computer Science Essay

Published: November 9, 2015 Words: 4665

Infrastructure less networks has found their significant place in the recent trends of engineering and technology. The research issues and challenges are paving the ways for the excellent research opportunities. In general, in wireless networking there are two main architectures: infrastructure (single-hop) networks and mobile ad hoc (multi-hop) networks. Even though users can handover between base stations or access points and roam among different networks, the conceit of mobility is limited within the coverage areas of the base stations or access points. MANET (Mobile Adhoc Network) denotes to a multi hop packet based wireless network collection of a set of mobile nodes that can have capability of communication and mobility at the same time, besides using any kind of fixed wired infrastructure. This network is really a self organizing and adaptive networks which can be formed and deformed on-the-fly without the essential of any centralized administration. In a MANET, available user's mobile devices are the network, since they must cooperatively provide the functionality usually provided by the network infrastructure (for e.g., routers, servers, switches). Most fundamental problems in adhoc networking is how to deliver data packets among the nodes capably lacking of predetermined topology or centralized control.

It is also important to provide the Quality of Service (QoS) when the protocols are used in communication. In this thesis, we proposed and implemented a QoS based framework for MANET for reliable communication.

MANET Architecture

Mobile Ad hoc network (MANET) is sometimes called a mobile mesh network which stands a self configuring network of mobile nodes associated by wireless links. The nodes are dynamic and arranged in the form of mobile topology. Each device in the MANET is open to move independently in any direction leads to change its link to other devices. Fig 1 shows the architecture of MANET.

A network without any base station is said to be infrastructure less network or multi-hop. MANET is a multi hop packet based wireless network which is a collection of two or more devices connected with network capability. Mobile ad hoc network supports the computation through anytime anywhere. There are two topologies in an ad hoc network.

Heterogeneous

Homogeneous

Fig 1.1. Heterogeneous structure of MANET

Fig 1.2. Homogeneous structure of MANET

Figures 1.1 and 1.2 shows the homogeneous and heterogeneous architecture of ad hoc network. Heterogeneous topology of network possesses difference in capabilities while Homogeneous topologies of the network, all the nodes have identical capabilities and responsibilities.

MANET USAGE AREA

The MANET is widely used in many areas:

Military situation - MANET supports strategic network for automated battle fields and military communications.

Rescue process - It present tragedy recovery which is a substitute of predetermined infrastructure network in case of environmental disaster.

Conferences.

Sensor networks - It consists of devices that have the capability of sensing, task of computing and wireless networking.

Free internet connection sharing - It also allows us to share the internet with others.

Mobile devices.

Students on campus.

Data network - used for the swapping of data among the mobile devices.

Device network - Device Networks supports the wireless connection among the various mobile devices so as to they can communicate.

TYPES OF MANET

Mobile ad hoc networks can be classified as:

Vehicular ad hoc network - Provides communication among roadside equipment.

Internet based mobile - These are one form of ad hoc networks which link mobile devices and static internet-gateway nodes.

Intelligent vehicular ad hoc network - It is a type of artificial intelligence which helps vehicles to perform in intelligent behaviors during vehicle to vehicle crashes, drunken driving, and accidents.

CHARACTERISTICS OF MANET

Mobile Ad hoc Network is a set of independent mobile nodes which can communicate with each other via radio waves. However, the mobile nodes that are in transmission range of other can directly communicate, instead of others need the support of intermediate nodes to route their packets. The structure of this network are fully distributed, it can work at every place without the help of any type of infrastructure. Utilization of these networks is highly flexible and robust. The features of these networks are summarized as follows:

Nodes can accomplish the roles of both hosts and routers.

Dynamic network topology facilitates frequent routing updates.

Communication via wireless means.

No centralized controller and infrastructure.

Can be set up anywhere.

Autonomous, means no infrastructure needed.

Energy constraints.

Generally, mobility nature of communication terminals makes the topology of the distributed network time varying. Furthermore, the dynamical nature of the network arrangements increases the challenges of the scheme of ad hoc networks. Individual radio terminal (RT) is usually powered by an energy limited power source (taken as rechargeable batteries). Consumption of power of every radio terminal might be divided generally into three parts as follows:

Power consumption for data processing within the RT.

Power consumption to transmit its specific information to the destination.

Power consumption once the RT is used as a router, which is forwarding the information to a different RT in the network.

The most critical issue in the design of ad hoc networks is the energy consumption. The mobile devices generally have limited storage and small computational capabilities. Since, they deeply depend on other hosts then resources for data access and information handling. Reliability in network topology must be guaranteed through efficient and secure routing protocols for Ad hoc networks. Table 1.1 shows the most important characteristics of MANET.

Important characteristics of MANET

Description

Energy Constraints

may rely on batteries or other exhaustible means

for their energy. The most important system

design optimization criteria may be energy

conservation for these nodes.

Mobile wireless networks are generally more

prone to physical security threats than fixed

cable nets. Possibility of eavesdropping,

minimization of denial-of-service type attacks

should be considered carefully.

Dynamic Topologies

Nodes are free to move arbitrarily with different

speeds and thus the network topology may

change randomly and at unpredictable times.

Limited Bandwidth

Wireless links continue to have significantly

lower capacity than

infra structured

networks.

Additionally, the realized throughput of wireless

communication after accounting for the effects

of multiple access, noise, fading, and

interference conditions is often much less than a

radio's maximum transmission rate.

Table 1.1. Important characteristics of MANET

APPLICATIONS OF MANET

The field of wireless networking emerges from the combination of cellular technology, personal computing, and therefore the web. The actual fact is because of the increasing interactions between communications and computing, access info every place at any time. Commercially massive kind of networks exist, this can be ranging from the well-known arrangement of cellular networks towards non-infrastructure wireless ad-hoc networks.

Technological improvements within the mounted wireless network on-the-fly networks or wireless ad-hoc area unit characterized by the shortage of infrastructure. Characteristics of nodes in a mobile ad-hoc network unit free to move and establish themselves in an absolute fashion. Thus every user is liberal to roam concerning whereas communication with others. There exists a path between every pair of the users could have multiple links, and conjointly the radio between them are often heterogeneous. It permits an association of several links to be a district of an equivalent network. Specifically, Mobile ad-hoc networks will operate in a complete fashion or may probably be connected to a bigger network like the net.

In many situations, Ad-hoc networks are suited for use where an infrastructure is unavailable or to arrange one is not cost effective. Some of various possible benefits of mobile ad-hoc networks is for some business environments, need for collective computing might be more significant outside the office environment than inside, like involving in a business meeting external to the office to short clients on a particular task. Process of work has been processing on to introduce the basic ideas of game theory and its uses in communications. This Game theory initiates from economics and has been applied in various fields.

A mobile ad-hoc network can similarly used to organize for disaster recovery operation, since the entire communication infrastructure is destroyed and restoring communication is also crucial. With the help of MANET, the infrastructure could be set up very easily in hours.

In an educational application, by using MANET virtual classrooms or conference rooms were setup. Also facilitates ad hoc communication environment during conferences, lectures and meeting.

MANET ROUTING PROTOCOLS

Mobile ad hoc network's routing protocols are typically subdivided into three main categories like Proactive protocols, Reactive routing protocols and hybrid routing protocols. Fig 1.3 shows the Routing Protocols in Mobile Ad hoc network.

Hybrid

Protocols

Proactive

Protocols

Reactive

Protocols

MANET

Routing Protocol

AODV

DSR

ACOR

ABR

WRP

OGSR

ARPAM

OLSR

ZRP

DSDV

TORA

OORP

DYMO

FSR

Fig 1.3.Routing Protocols in MANET.

The efficient routing protocols can provide important benefits to mobile ad hoc networks which in terms of both performance and reliability.

REACTIVE ROUTING PROTOCOLS

Reactive MANET protocols search a route towards the destination node after there is a necessity to send data. Initially, sender will begin by transmitting route requests in the entire network. The source node will remains for the destination node or an intermediate node to reply with a list of intermediate nodes among the source and destination. Communication in this way is known as the global flood search, possibly brings about significant delay formerly the packet can be transmitted. These protocols also require the transmission of a significant quantity of control traffic. Hence, the reactive MANET protocols are most suitable for networks with high node mobility or wherever the nodes transmit data rarely.

The following are the reactive protocols:

Ad hoc on demand distance vector (AODV)

Dynamic MANET On demand (DYMO)

Dynamic source routing (DSR)

Associatively based routing (ABR)

Admission control enabled on demand routing (ACOR)

SourceCharacteristics

Destination

Routing Information

On demand

NodeSecurity Threats

Fig1.4. On demand routing protocol

If on demand routing protocols are used, whenever data is routed towards a destination, it requires a source node to initiate a search for the destination. When the network is large, there occurs significant latency before the destination is found. The advantage of employing this protocol in a network environment is as follows:

The Adhoc on-demand Distance Vector routing protocol contains a numerous of messages which it utilizes for route discovery, neighbor detection, route maintenance and repair.

There is a number of implementation AODV available for the Linux platform.

The main drawback in this protocol is the improvement of ad-hoc routing protocols which has been deliberate because of the present operating systems do not offer adequate straight system services for their execution.

PROACTIVE ROUTING PROTOCOLS

These protocols are table driven which actively determine the layout of the network. Through the continuous and consistent interchange of network topology packets among the nodes in the network, also a whole picture of the network is preserved at every single node. This in turn leads to slight delay in defining the route to be taken. This criterion is especially important for time-critical traffic. On the other hand, the drawback to a proactive MANET protocols exists that the life span of a link remains significantly short. However, this phenomenon is carried about through the improved mobility of the nodes, which in turn extract the routing information in the table worthless quickly. By analyzing overall, proactive protocols performs best in networks that have small node mobility before the nodes transmit data frequently. The following are the proactive routing protocols.

Destination Sequence Distance Vector (DSDV)

Optimized Link State Routing (OLSR)

Fisheye State Routing (FSR)

Cluster Head Gate Way Switch Routing (CGSR)

Wireless Routing Protocol (WRP)

Proactive

Routing

Protocol

Destination

Routing Information

Node

Fig 1.5. Table driven protocol

When proactive routing protocols are engaged a node would acquire routing information to a destination previously it would need to route data so as to destination. Routing tables are maintained for this purpose. Route updates are switch over occasionally to reflect the changes in topological information. The proactive routing protocols usually require the maintenance of routing tables, since the mobile ad hoc network changes dynamically. Furthermore, swapping of route updates would consume bandwidth and messages could contribute to a major amount of overhead. The advantages of using this protocol in mobile ad hoc network are as follows:

Access up-to-date network - By assessing the known and attempting to discover new routes updated, they try to maintain most recent network.

Efficiency - The protocol efficiently forward packets, since the route is known when the packet arrives at the node.

HYBRID ROUTING PROTOCOL

An approach both proactive and reactive routing protocols every work finest in oppositely dissimilar scenarios, valid reason and chance to develop hybrid routing protocols ( combination of both proactive and reactive). However, these hybrid protocols can be used to search a stability among the proactive and reactive protocols. The abstract plan behind the hybrid routing protocols is to usage of the proactive routing mechanisms in certain areas of the network at definite times and reactive routing applied for the rest of the network. The operation done through proactive routing is limited to a lesser domain in order to diminish the control overheads and delays. Protocol of this type is used to locate the nodes outside this domain, because of more bandwidth efficient in a regularly changing network.

The following are hybrid routing protocols:

Zone routing protocol (ZRP)

Order one routing protocol (OORP)

Temporary ordered routing algorithm (TORA)

MANET features

Mobile ad hoc network possesses the following features:

Distributed operation:

Subsequently there is no background setup for the central control of the network methods, both the management and control of the network is disseminated among the terminals. Moreover, the nodes concerned in a MANET would collaborate amongst themselves and every node turns as a relay when needed, which in turn leads to implement functions such as security and routing.

Autonomous terminal:

In MANET, each and every mobile terminal is an autonomous node, it act and function as a host and a router. Put differently, besides the basic processing aptitude as a host, the mobile nodes which also perform switching purpose as router. Hence, endpoints and switches are usually indistinguishable in MANET.

Multi hop routing

The fundamental types of adhoc routing procedures can be split into single-hop and multi hop, on the basis of different link layer attributes and routing protocols. However, Single-hop MANET is effortless than multi-hop in terms of configuration and execution, in association with the cost of smaller functionality and applicability. When data packets are delivering after a source to its destination available to the direct wireless communication range, available packets must be sent through one or more intermediate nodes.

Fluctuating link capacity

The nature of high bit error rates of wireless structure strength be more profound in a MANET. Several sessions shares the one end-to-end path. The channel over which the terminals communicate depends on to noise, interference and fading and has a lesser amount of bandwidth than a wired network. Sometimes the path among any pair of users which traverse multiple links while the link themselves can be heterogeneous.

Dynamic network topology

Since the nodes are mobile in MANET, the topology followed in network may change rapidly, randomly and the connectivity among the terminals may differ with time. Design principles in MANET should adjust to the traffic, the propagation environments and the mobility prototype of the mobile adhoc network nodes. Mobile ad hoc network form their own network on the fly by which the mobile nodes can dynamically create routing among themselves as they change about. Furthermore, a user in the MANET could not only function within the ad hoc network, necessity occurs sometimes that require access to a public fixed network (for e.g. Internet). In some scenarios, the MANET nodes are mobile devices with a lesser amount of CPU processing capability, low power storage and small memory size. Such devices need optimized algorithms, mechanisms that process the computing and communicating functions.

UNIPATH ROUTING IN MANET

Routing protocols are used to detect and preserve the routes among the source and the destination nodes. As mentioned earlier in section 1.2, there are two main classes of adhoc routing protocols which include routing table and on-demand protocols. In table driven protocols, each and every node preserves a routing table consisting routes to the entire nodes in the network. Nodes participated in network must occasionally swapping messages with routing information to remain routing tables up-to-date. Hence, the routes between nodes are computed and stored, besides when they are not needed. Since the vibrant nature of ad hoc networks, there is an option for substantial amount of routing data's may have to be swapped in order to maintain routing information accurate.

In case of on-demand protocols, all participated nodes can compute routes only when they are needed. On this feature, on-demand protocols are more scalable to vibrant, and highly suitable for large networks. In particular, whenever a node requires a route to an additional node, it then begins a route discovery process to detect a route.

There are two main phases in On-demand protocols.

1. Route discovery - It is the process of detecting a route among two nodes.

2. Route maintenance - It is the method of repairing a busted route or detecting a novel route in the occurrence of a route failure.

Some of the majority presently planned routing protocols for ad hoc networks which are unipath routing protocols. Normally, unipath routing requires simply a single route is used among a source and destination node. Dynamic Source Routing and Adhoc On-demand Distance Vector protocols are the two most widely used unipath routing protocols..

DYNAMIC SOURCE ROUTING

DSR is an on-demand routing protocol meant for ad hoc networks. Compare with any source routing protocol, in case of DSR the source includes the complete route in the packets header. However, the intermediate nodes make use of this to sent packets towards the destination and keep a route cache including routes to other nodes. Following are the two phases:

Route discovery:Consider a scenario if the source node does not having a route towards the destination in its route cache, then it broadcasts a route request (RREQ) message identifying the destination node intended for which the route is requested. The information encrypt in this RREQ message contains a route record which identifies the succession of nodes pass through by the message. In a case, when an intermediate node obtains a RREQ, it then checks to see if it is previously in the route record. If record presented already, it drops the message. This process is done to prevent routing loops. The intermediate node sends the RREQ to the succeeding hop according to the route particular in the header. Afterwards, when the destination receives the RREQ, it then sends back a route reply message. Whenever, the destination has a route towards the source in its route cache, later it can forward a route response (RREP) message beside this route. In other words, the RREP message can be directed next to the reverse route back to the source. For instance, middle nodes may also utilize their route cache to reply to RREQs.

Route maintenance: Analyzing this scenario, when a node finds a broken link while trying to send a packet to the next hop, this can forward a route error (RERR) message and send to the source enclosing the link in error. The time when an RERR message is received, all other routes having the link in error are detached with the intention of node.

AD HOC ON-DEMAND DISTANCE VECTOR(AODV)

AODV is an on-demand routing protocol in adhoc networks. As compared to DSR, that uses source routing protocol, AODV uses hop-by-hop routing through preserving routing table entries at middle nodes in the network.

Route Discovery: The route discovery method is commenced while a source requires a route to a destination. The route discovery can be initiated, by means of source floods the network by means of a RREQ packet identifying the destination for requested route. Sequentially, when a node receives an RREQ packet, the node verifies to notify whether it is the destination or path to the destination. Whenever, an either case is true, RREP packet is generated by the node, which is thrown reverse to the source beside the turnaround the path. However, each node along the reverse path makes a forward pointer towards the node it obtained the RREP. This entire process placed up a forwarded path beginning the source to the destination. Scenario likes if the device is not the destination and does not have a correct path towards the destination, and then it rebroadcasts the RREQ packet. Duplicate RREQ packets are discarded at the intermediate nodes. When the source node receives the primary RREP, then it can start on sending data to the destination. Each and every entry in the node routing table and the entire RREQ and RREP packets are labeled with a destination succession number to decide the relative degree out-of-datedness of routes. If there are number of destination than the sequence number pointed to a more current route.

Route Maintenance: From time to time a node detects a busted link while attempt to forward a packet to the following hop, such time it produces a RERR packet which is send to the entire sources using the busted link. However, this RERR packet erases all routes by means of the link beside the way. On the other hand, if a source obtains a RERR packet while a path to the destination is still essential, then it initiates a new route discovery process.

MULTIPATH ROUTING IN MANET

Standard routing protocols in ad hoc wireless networks like AODV and DSR, which are mostly planned to find out a single route among a source and destination node. To discover numerous routes among a source and destination node multipath routing is used. These multiple paths among source and destination node pairs can be utilized to recompense for the vibrant and random movement of ad hoc networks.

BENEFITS OF MULTIPATH ROUTING

As mentioned above, multiple paths can provide load balancing, higher aggregate bandwidth, and fault-tolerance. Load balancing can be accomplished by spreading the traffic along various routes. However, it can alleviate congestion and bottlenecks. In a perspective of fault tolerance, multipath routing can provide route resilience. Because of bandwidth limitation in a wireless network, the possible routing beside a single path may not offer sufficient bandwidth for a connection. Furthermore, there exist more bandwidth available, achieves smaller end-to-end delay. Evaluation results show so as to using multipath routing in ad hoc networks of large density consequences in improved throughput than using unipath routing.

ISSUES IN MOBILE AD HOC NETWORK

Scalability is one of the most significant open problems. The problem of scalability in adhoc networks can be generally defined as whether the network is capable to provide an adequate level of service to packets still in the occurrence of a huge number of nodes in the network. In case of wired network, such capability is closely related as to how rapidly network protocol control overhead enlarges as a purpose of a raise in the number of nodes while link changes.

Ad hoc network do not assume the accessibility of a fixed infrastructure, since it follows individual nodes may have to rely on limited power sources and portability. The capability of fixed and wireless networks to satisfy quality of service (QoS) requirement is an important problem in ad hoc networks.

PROBLEM DEFINITION

MANET is a difficult task to define, however most important challenges in MANET are to provide quality of service in multimedia. Through resource limitation, contention for bandwidth, time varying links, etc. QoS is hard for ad hoc environment. Quality of service is usually defined as a set of service requirements that needs to be met by the network at the time of transporting a stream of data from source to destination. Expectation on network topology to guarantee a set of measurable service attributes to the user in terms of reliable communication through increase throughput, packet delivery fraction and decrease end to end delay.

Reliablity

Energy constraint

Network lifetime

Link failure

Delay

EXISTING METHODOLOGY

On-demand routing protocols in particular, are widely studied because they consume less bandwidth than proactive protocols. Ad Hoc On-demand Distance Vector and Dynamic Source Routing and Dynamic Manet On-demand are the three most frequently studied and analyzed on-demand ad hoc routing protocols. Previous literatures have shown certain limitations of the three protocols. The main reason is that all three of them are built and rely on a unipath route for each data session. These three routing protocols have to invoke a route discovery process at the time of link break on the active route.

On-demand multipath routing protocols can alleviate these problems by establishing many paths between a source and a destination in a single route discovery. Discovery of new route is invoked only when all of its routing paths and lack of security and also produce the non reliable communication. Plenty of existing work relates the combination of proactive and reactive routing protocols like SHARP, ZRP etc. But the most important issue in MANET environment is to provide the Quality of Service (QoS) when the protocols are used in communication. In this thesis, we proposed and implemented a QoS based framework for MANET for reliable communication. The proposed scheme is to operate Consecutive routing process (CRP), the three routing protocol in the same network. Background of the network is chosen as reactive protocol. Whereas the reactive, proactive and hybrid protocols are operate in same network. FSR, DYMO, ZRP are the proactive, reactive and hybrid protocols used in our proposed scheme. The design of the framework is also supported by the coded algorithm.

OBJECTIVE OF THE THESIS

The objective of the thesis is to bring out Quality of Service (QoS) mobile ad hoc network environment for reliable communication. The specific objective of the thesis is to increase throughput, packet delivery fraction and decrease end-to-end delay of the network scenario under consideration. Furthermore, the framework also provides the following:

Communication reliability is achieved using the newly devised algorithm, which allows the repeated interactive and non-interactive simulation runs for various node densities.

Utilization of Consecutive routing protocols in the network renders most prominent performance.

Reliable communication check is done with the help of proposed framework which follow the proper request, discovery and maintenance of route.

High delivery with respect of both packet delivery fraction and throughput is accomplished through this framework.

The algorithm allows faster understanding of node communication of any specified network conditioning.

MAJOR CONTRIBUTIONS

The main focus of this research is to make meaningful involvement in the improvement of QoS in MANETs in application layer. The thesis has made the following contributions in effective way to achieve reliable communication in MANET environment.

EDYMO which deals with the energy constraint for reducing the energy consumption, increasing life time and deliver better performance

Network set up module consists of IEEE 802.11 DCF (Distributed Co-ordination Function) for MAC layer, Random Way point Model (RWM) for Mobility variations and Random Early Detection (RED) for congestion control purposes.

Routing module consists of Consecutive routing protocol study of EDYMO, FSR and ZRP routing protocols which are significant for bringing out reliable communication standards for network scenario. The corresponding methodologies were also included in the linked study phase for each protocol individually.

Performance analysis module which is a group of QoS metrics for better performance of network scenario, reliability for communication ease and security to avoid malicious attacks.

ORGANIZATION OF THE THESIS

The work reported in this thesis is organized into six chapters. In Chapter 1, the background of MANET, the objectives are discussed and explain the challenges and problem definition of MANET and literature survey of manet and related works. In Chapter 2, discussion about various routing protocols important for the analysis necessary for the proposal and an algorthmic approach for the design of QoS based Mobile ad hoc network framework for reliable communication is presented. In Chapter 3, analysis and comparison the results and describtion on the major contributions of this research work and the scope for future work are summarized. Illustrative examples are included in each chapter to demonstrate the effectiveness of the proposed framework