Download Mobile Adhoc Networks Progress-Computer Sciences Applications-Project Report and more Study Guides, Projects, Research Applications of Computer Sciences in PDF only on Docsity!
i
Table of Contents
Objective of the Report ...................................................... Error! Bookmark not defined.
- Chapter 1:Background & Related Work Abstract ii
- 1.1 Routing Protocols..........................................................................................................
- 1.1.1 Categories
- 1.1.2 Differences
- 1.2 Well known routing algorithms
- 1.2.1. DSDV....................................................................................................................
- 1.2.2. TORA....................................................................................................................
- 1.2.3. DSR
- 1.2.4. AODV
- 1.2.5 Global State Routing Protocol
- 1.2.6 Fisheye State Routing Protocol
- 1.3 Summary
- Chapter 2:Research Trends & Important issues in MANETs
- 2.1 Introduction
- 2.2 Previous Literature Work
- 2.3 Analysis Methodology
- 2.3.1 Major MANET Issue Categories
- 2.3.2 Literature study methodology
- 2.4 Important Parameters of MANETs
- 2.4.1 Simulation Metrics for MANETs
- 2.5 Conclusions
- Chapter 3:Network Simulator
- 3.1 Network Simulator (NS)
- 3.1.1 Marc Greis Tutorial..............................................................................................
- 3.1.2 NS by Example
- 3.1.3 NS Manual, NS Search and NS Mailing List
- 3.1.4 Network Animator (NAM)
- 3.2 Simulations
- Bibliography
ii
Abstract
Mobile ad-hoc networks, also known as short-lived networks, are autonomous systems of mobile nodes forming network in the absence of any centralized support. This is a new form of network and might be able to provide services at places where it is not possible otherwise. Absence of fixed infrastructure poses several types of challenges for this type of networking. Among these challenges is routing.
By routing, we mean process of exchanging information from one station to the other stations of the network. Routing protocols of mobile ad-hoc network tend to need different approaches from existing Internet protocols, since most of the existing Internet protocols were designed to support routing in a network with fixed structure. In the academic and industrial world, those who think about such things have written quite a few papers proposing various routing solutions for mobile ad-hoc networks. Proposed solutions could be classified into six types: table-driven, on-demand, hierarchical, power- aware, geographical, and multicast protocols.
1.1.2 Differences
Some of the differences between Table-driven (pro-active) and on-demand (reactive) routing protocols are:
Pro-active Reactive Attempt to maintain consistent, up-to-date routing information from each node to every other node in the network.
A route is built only when required.
Constant propagation of routing information periodically even when topology change does not occur.
No periodic updates. Control information is not propagated unless there is a change in the topology.
Incurs substantial traffic and power consumption, which is generally scarce in mobile computers
Does not incur substantial traffic and power consumption compared to Table Driven routing protocols. First packet latency is less when compared with on-demand protocols
First-packet latency is more when compared with table-driven protocols because a route need to be built A route to every other node in ad-hoc network is always available
Not available.
Table 1: Differences b/w Reactive and Pro-active
1 .2 Well known routing algorithms
1.2.1. DSDV
DSDV [1] performs quite predictably, delivering virtually all data packets when node mobility rate and movement speed are low, yet failing to converge as node mobility increases. We found that DSDV works better when it is used without link layer feedback. For that reason, it can be argued that DSDV is treated unfairly in this study, because all
the per-data-packet acknowledgments sent by the MAC protocol could be treated as routing overhead that should be charged against the other protocols. For an environment that has a rate of topology change on the order of the 1m/s data shown here (i.e., 1 link status change per second), DSDV would actually have a lower routing overhead than either TORA, DSR, or AODV while providing an equivalent packet delivery ratio. This effect occurs because the routing overhead would be dominated by the number of per- data-packet acknowledgments, Which increases with the number of data packets sent? However, as described in Section 7.4, even when stationary, the nodes in a real ad hoc network see significant rates of topology change due to wireless propagation effects and multipath resulting from objects in the world moving around the nodes. These real-world effects would make the use of DSDV risky in a real ad hoc network.
1.2.2. TORA
Although the worst performer in our experiments in terms of routing packet overhead, TORA[2] still delivered over 90% of the packets in scenarios with 10 or 20 sources. At 30 sources, the network was unable to handle all of the traffic generated by the routing protocol, and a significant fraction of data packets were dropped in these scenarios. Packet losses in TORA resulted from two main sources. The first is short-lived routing loops caused by the delay between a link reversal at one node and the link-reversals at neighboring nodes. The second, and more serious one, is packet loss due to congestion caused by a positive feedback loop in the behavior of the IMEP reliable broadcast protocol used by TORA to distribute its routing updates. Given the increased probability of loss for broadcast packets in wireless networks, our experience with TORA/IMEP argues that routing protocols must be designed to be tolerant of the loss of their broadcast packets, rather than attempting to eliminate the loss via a reliable broadcast algorithm.
1.2.3. DSR
The performance of DSR [3] was very good at all mobility rates and movement speeds we studied. The analysis of routing overhead bytes, however, shows the high cost of
topology table and broadcasts it to its neighbors. The routing table information is exchanged periodically with the neighbors only.
1.2.6 Fisheye State Routing Protocol
FSR [6] protocol is an extension of GSR protocol. It attempts to reduce the size of update messages in GSR without seriously affecting the routing accuracy. The reduction in routing size is obtained by using different exchange periods for different entries in the routing table. Entries corresponding to nodes within the smaller scope are propagated to the neighbors with the highest frequency. For example, entries corresponding to nodes within a single hop count are propagated more frequently, within two hop count are propagated less frequently, and others are propagated even less frequently. As a result, a considerable fraction of link state entries are suppressed, thus reducing the message size. This strategy results in routes to destinations whose accuracy decreases as their distance from the node increases. However, as the packet gets closer to the destination, the route becomes progressively more accurate.
1 .3 Summary
The routing protocol is one of the fundamental protocols in MANETs. Standard routing protocols for MANETs have not yet been defined. Currently, there are four leading routing protocols, AODV, DSR, OLSR, and TBRPF, as determined by the IETF MANET group [7]. Before our research, most prior work focused on simulation-based comparisons among different MANET routing protocols. However, due to a lack of proper characterization of different MANET protocols, these simulation experiments are not well designed. For example, the simulation results from different research groups cannot be directly compared [8]. There are no clear conclusions that can be drawn from this prior work. In other words, the relationship between the simulation conditions and MANET routing protocols remains unclear. Therefore, the conclusions based on the simulation experiments cannot be generalized and new methodologies to study MANET routing protocols are clearly needed.
Chapter 2
Research Trends & Important issues in
MANETs
2.1 Introduction
A mobile ad-hoc network (MANET) is based on a self-organizing and rapidly deployed network. MANET applications include supporting battlefield communications, emergency relief scenarios, law, enforcement, public meeting, virtual class room and other security-sensitive computing environments. The ad-hoc networking technology has stimulated substantial research activities in the past 10 years. Many scholars were attracted to investigate this domain for further research and learning. Numerous problems and challenges exist in this field because of the frequent and unpredictable MANET topology changes.
Hundreds of on-going research issues are deployed involving widespread situations for MANETs If the researchers begin to study the ad-hoc network, they usually need to spend a lot of time collecting and arranging related literature. Much work is then needed to find the research trends and directions. Macroscopic information and research trends would be very helpful for novices. This could increase the research efficiency for beginners in this field and help them to search the applicable research direction rapidly.
Researchers usually subjectively focus on research aspects. This causes researchers to ignore the characteristics behind the issues. If they have not considered all of the characteristics involved in each issue, they will not examine the implicit problems. If we can provide the appropriate characteristics of each issue to the researchers, they could better understand the characteristics of each issue and investigate potential strategies.
Addressing, bandwidth management, security, fault tolerance, QoS and multimedia, standards and products. They only provided an overview for limited issues in this field Researchers will have difficulty obtaining comparisons between the issues from these literature.These surveys provided a technical overview and potential applications for a specific MANET project.
The Terminodes project [11] is a 10-year joint effort by seven Swiss research institutes begun in 2000. Their aim is to study and prototype a large-scale ad-hoc network with the emphasis on the self-organization feature. The project has higher-level abstraction goals like the OSI seven layer network design.
Ramanathan [12] discusses some open OSI seven layer problems. The ad-hoc Networking is a multi-layer problem. For example, a similar multi-layer issue is security in MANETs. Some newly emerging challenges are MANET designs that can take advantage of new hardware technologies. Most existing solutions for saving energy in MANETs revolve around the power reduction by the radio transceiver.
These papers cover current research issues and describe major technical challenges, including networking, real-time services, and software. However, they do not present the numerical analysis data to provide advanced information for various techniques.
2.3 Analysis Methodology
This section presents our analysis methodology. Section 2.3.1 introduces some major issues and sub-issues involving MANETs. Various literature and searching techniques are described in Section 2.3.2. The literature collection and a detailed description of the searching principles used in this work are described.
2.3.1 Major MANET Issue Categories
In this section the MANET research issues are presented and classified. Hundreds of research aspect have been developed and discussed in this field. To analyze various research issues, this article covers most of the major investigation problems. Various Fundamental and frequently discussed aspects of MANETs are identified and grouped into fifteen categories. These issues have the potential to significantly increase MANET survivability: (1) Routing: Routing is an essential protocol in this field, because changes in network topology occur frequently. An efficient routing protocol is required to cope with highly fluid network conditions. (2) Multicasting/ Broadcasting: Multicast service supports users communicating with other members in a multicast group. Broadcast service supports users communicating with all members on a network. (3) Location Service: Location information uses the Global Positioning System (GPS) or the network-based geo-location technique to obtain the physical position of a destination. (4) Clustering: Clustering is a method to partition the hosts into several clusters and provide a convenient framework for resource management, routing and virtual circuit support. (5) Mobility Management: In the ad-hoc network environment, mobile hosts can move unrestricted from place to place. Mobility management handles the storage maintenance and retrieval of the mobile node position information. (6) TCP/ UDP: TCP and UDP are the standard protocols used in the Internet. Data applications running over MANETs, such as http and real audio need transport layer protocols like TCP and UDP to send packets over the links. (7) IP Addressing: One of the most important issues is the set of IP addresses that are assigned to the ad-hoc network. IP addressing and address auto-configuration have attracted much attention in MANETs.
Each issue covers a wide range of sub-issues. In the routing issue, the sub-issues were routing protocol, and proactive/ reactive/ hybrid schemes, etc. New keywords were selected as comprehensive as possible to represent the scope of each issue. The literature data source in this work was TM collected from IEEE Xplore. It contains more than 610,000 articles over 12,000 individual publications TM since 1988 [13]. The IEEE Xplore electronic resource database provides a powerful and convenient interface for searching. Papers could be located by specifying one or more titles, issues, or other criteria based on the keyword ad-hoc networks.
2.4 Important Parameters of MANETs
In this section, a detailed description of the simulation metric factors is given. When the performance of algorithms or approaches is evaluated, metrics play an important role. Different metrics are adopted in different studies. A number of performance metrics are used to quantify the differences. The top five simulation metrics are then presented.
2.4.1 Simulation Metrics for MANETs
When researchers develop a new algorithm or method, they can use the metrics to prove that their approach is better than others. To represent the metric characteristics, some common simulation variables are listed as follows: (1) Mobility: The mobility metric is used to measure movement in the network by calculating the relative node movement between all pairs of nodes in the network. (2) Overhead: The overhead metric means how many extra messages were used to achieve acceptance rate improvement. (3) Transmission Range : the transmission range metric is limited by the power constraints, frequency reuse and channel effects. It represents the distance the beacon message must travel.
( 4) Packet Delivery Ratio: The packet delivery metric ratio presents the ratio between the number of sent packets from the application layer and the number of received packets at the destination nodes. (5) Throughput: The throughput metric is calculated by dividing the total number of packets sent by the time the first packet is received minus the time the last packet is received.
As described above, these metrics are important because they can be used to identify the properties for design methods and provide valuable information about simulation results. Based on these results, researchers can use these metrics to evaluate their proposed methods when they process a simulation. The overhead simulation metric is the most frequently used factor in the Y axis and mobility is the most frequently used factor in the X axis. The reasons for this phenomenon are listed as follows. (1) MANET features include low bandwidth, limited storage space and low computing capacity. If the overhead for a researchers proposed method is high, the method cannot work well in MANETs. Therefore, the overhead is a very important simulation metric when researchers compare their methods with other existing methods. (2) Mobility is the main characteristic in the MANETs. Therefore, researchers must use this metric to evaluate their proposed methods.
2.5 Conclusions
In this study, MANET related papers within the most recent years on IEL, were collected and our observations were summarized. Possible research directions and trends in this field were proposed. To determine the research trends in MANETs, the context of 1, papers were sorted into fifteen issues and analyzed in detail. According to our analysis results, the routing and power management issues have grown very fast and were the most popular in recent years. Although the IP addressing and fault tolerance issues have not been discussed very often, they will have potential study value in the future. Based on
Chapter 3
Network Simulator 2
In this chapter, my main focus will be on different simulations performed in Network Simulator 2 environment from very basic to required level. I will also touch the issue of problems and error occurred in NS2 installation and simulations.
As mentioned earlier in the text, Network Simulator 2 [14] is used as the simulation tool in this project. NS was chosen as the simulator partly because of the range of features it provides and partly because it has an open source code that can be modified and extended.
There are several different versions of NS and at the current time the latest version is Ns-2.29 while Ns-3.0 and Ns-3.1 is under development. The latest version of ns has been Used in this study .This chapter describes the simulation environment in Section 3.
3.1 Network Simulator (NS)
Network Simulator (NS) is an object-oriented, discrete event simulator for networking research. NS provides substantial support for simulation of TCP, routing and multicast protocols over wired and wireless networks [14]. The simulator is a result of an on-going effort of research and development.
NS is written in C++, with an OTcl interpreter as a command and configuration interface. The C++ part, which is fast to run but slower to change, is used for detailed protocol implementation. The OTcl part, on the other hand, which runs much slower but can be changed very quickly, is used for simulation configuration. One of the advantages of this split-language programming approach is that it allows for fast generation of large scenarios. To simply use the simulator, it is sufficient to know OTcl. On the other hand,
one disadvantage is that modifying and extending the simulator requires programming and debugging in both languages simultaneously.
3.1.1 Marc Greis Tutorial
The very first thing to do for a new user of NS is to read MarcGre is tutorial. There is a Link to this tutorial on the webpage of NS which can be found at www.isi.edu/nsnam/ns/.
The purpose of this tutorial is to make it easier for new NS users to use NS and NAM, To create their own simulation scenarios for these tools and to eventually add new functionality to NS.
3.1.2 NS by Example
On the web page of NS, there is a link to another tutorial for NS. This tutorial has been written by Jae Chung and Mark Claypool and its purpose is to give new users Some basic idea of how the simulator works, how to setup simulation networks, where to look for further information about network components in simulator codes, how to create new network components and so on. In particular, it explains the linkage between the two languages used in NS, namely C++ and OTcl. One can find some very good examples and brief explanations in this tutorial, which is the second tutorial to study after reading MarcGreis tutorial.
3.1.3 NS Manual, NS Search and NS Mailing List
In the NS Manual [14] one can find the answer to many questions. A link to this Manual can be found on the web page of NS. However, if no answer can be found in the Manual, the NS mailing list archives should be searched. The archive keeps all previous emails sent to the ns-users mailing list. The ns-users mailing list should be
3.2 Simulations
Simulation Language:
set ns[new Simulator] set nf[open out.namw] proc finish{} { global ns nf $ns flush-trace close $nf exec namout.nam& exit 0 } $ns at 5.0 "finish" $ns run
Simulation 1. (Simple Wireless Network Simulation prototype 1)
Scenario:
Two fixed nodes moving within 600m x 600m flat topology DSR ad hoc routing TCP and CBR traffic Receiver move in and out of range
This model is working based on IEEE 802.11 MAC protocol. We set the routing protocol to DSDV (Destination-Sequenced Distance-Vector, LinkLayer type to LL, the interface queue type to Queue/DropTail/PriQueue, physical type to a wireless network interface, and channel type to a wireless channel. The topography for this simulation is set to a 600*600 flat grid and there are two nodes. The distance between two nodes is
200m. In this simulation, I observe both cases of when two nodes are in transmission range and out of the transmission range.
NS2 Code:
#==============================================================
Define options
#============================================================== set val(chan) Channel/WirelessChannel ;# channel type set val(prop) Propagation/TwoRayGround ;# radio-propagation model set val(ant) Antenna/OmniAntenna ;# Antenna type set val(ll) LL ;# Link layer type set val(ifq) Queue/DropTail/PriQueue ;# Interface queue type set val(ifqlen) 50 ;# max packet in ifq set val(netif) Phy/WirelessPhy ;# network interface type set val(mac) Mac/802_11 ;# MAC type set val(nn) 2 ;# number of mobilenodes set val(rp) DSR ;# routing protocol set val(x) 600 set val(y) 600
set ns [new Simulator] #ns-random 0
set f [open 1_out.tr w] $ns trace-all $f set namtrace [open 1_out.nam w] $ns namtrace-all-wireless $namtrace $val(x) $val(y) set f0 [open proj_out0.tr w] set f1 [open proj_out1.tr w]
set topo [new Topography] $topo load_flatgrid 600 600
create-god $val(nn)
set chan_1 [new $val(chan)] set chan_2 [new $val(chan)]
CONFIGURE AND CREATE NODES
$ns node-config -adhocRouting $val(rp) -llType $val(ll) -macType $val(mac) \
