1. ABSTRACT:
This paper describes about mobile ad-hoc networks. In mobile ad-hoc networks the simulation tools plays a vital part. The main simulation tools used in MANETS are described in this paper elaborately. Various aspects of MANET routing protocols, quality of service and security issues of the simulated tools have been explained. Advantages and disadvantages of simulation tool software and test bed is explained.
2. INTRODUCTION:
Mobile Ad-hoc network in simulation tool have been explained. There are much simulation software available in the market. In this paper I have used OPNET, NS2 and GloMoSim to find the routing issues, Quality of Service and security in MANET. The advantages and disadvantages are more comfortable in OPNET when comparing with NS2. Test bed is explained and it has opposed to simulation tools. Thus in this paper MANET simulation tools are explained elaborately.
3. MANET:
Mobile Ad-hoc network are wireless networks and there are no infrastructures. There are no connections with base stations or an access point as of in local area wireless networks. it creates a own network through collaborations of user radios. To discover other radios user should turns on their radio and it follows a protocol to produce their own existence. The information collected is used for their own communication route. Manet is a challenging protocol for many networks. Distributed solutions are provided by protocols in the absence of base stations and access points in the central node. Network topology is more important in routing protocol Manet. For tracking the topology there will be effectiveness in protocol for the direct co-relation. There will be a control in wireless network for the capacity Rosen et al (2006).
4. SIMULATION TOOLS:
Simulation tools are used in less cost the simulation tools are used to test the subject, which is no longer enough. With this simulation tools the user can experiment the long term project virtually without investing the large amount of money. Implementing test bed is very costly. Comparing with test bed simulation tools are cheaper to implement. There are many simulation tools available in the market for the computing field. They are
4.1. SIMULATION TOOLS USED IN MANETS:
There is believability in the simulation studies of Manet. It involves in the research of simulation based on trustworthy. There are four area in simulation research for credibility.
5. OPNET:
User can build a simulation topology because it is the vast amount of elements in the network. The OPNET modeller 8.1.A is very large in the model library. Vendor models are available in the collection which makes a real-life device from many vendors which is available right now in the market. There would be no information specifically for this model. Because this model look likes the real device of the protocol which they support. Comparing with the performance parameters it is not closely related to the real devices. The IP process information contains the routing performance in default while working as parameters. If the delay occurs by the router, it won't be the actual delay induced from the router Broeck et al. (nd).
5.1. OPNET IN MANET:
There will be a critical situation in the scenario when the objects are generated in real-time simulation. To design a real time simulation tool or real time system many techniques in scheduling has been proposed. There will be a schedule in the task for the dead line. In advance the scheduling in real time for the task priorities is well known and that all related to the individual character in the control component.
To validate the router model in OPNET, let's choose a cisco 2621 router, in which it has two autosensing 10/100 Ethernet ports which helps to evaluate the performance.
The router has been built with a central processor which is responsible for all packet processing functions. In order to measure the performance we disable the router cache and load the router with well defined routing patterns and by checking how the router for the traffic. For comparing and evaluating the performance of router we need some utility to generate some fake traffic. I have decided to use RUDE and CRUDE measurement tool from which can measure the performance with the metrics like packet loss, throughput, packet delay and distribution.
The RUDE stands for Real Time UDP Data emitter, with this process we can flood the packets to create traffic with in a microsecond. Hence, the UDP packets have the payload of 20 bytes. In the same way we need a process to receive packets for that CRUDE is used which stands for Collector for RUDE. The receiving of packets is not accurate when compared to sending and crude takes the time of 200 microseconds.
5.2. QoS :
Quality of service supports wireless LAN from the proposal of IEEE 802.11 e. There are two new mechanisms accessed in hybrid Co-ordination function (HCI).
The QoS does not support the original 802.11 and thus the simulation model been built by the Opnet modeller. The two models are comparing the performance in 802.11. the new 802.11 e will work on the current models such as to design and schedule a new queuing algorithm in the wireless MAC layer for the improvement in QoS. Tere is a good support for the EDCA model in the simulation for the services like video and voice. The Opnet modeller does not have an HCCA model in the latest version. To support a new QoS model the Opnet should be designed and programmed in C++ (Opnet, 2003).
5.3. SECURITY:
In two decades the security problem in manet has been authenticated and studied extensively. The feasible methods are key distribution in asymmetric method and key management. In routing protocol there will a less attention. Public key infrastructure has been introduced by Hass and Zhou. Asymmetric mechanisms are validated in the threshold signature of cryptography. The security extension to 802.11 are added and allowed the model feature in power adaptation, new routing and path loss. OPNET has the implementation of OSPF2 from the standard library.
5.4. ADVANTAGES:
Parallel technique is used to implement the simulation for the time constraints to achieve the desirable path. Main simulation tool is used for the techniques in OPNET modeller as a potential. The parallel technique does not perfect the simulation in some cases.
5.5. DISADVANTAGES:
The real time constraints are the parallel simulations and perfect technique. It is not made as parallel-safe in the nature of 802.11. The simulation performance is a problem for static window size. So that there will be no fall of two windows within the dependent (event Vu et al. N.d.).
6. NETWORK SIMULATOR ns-2:
Ns2 is a networking research. It supports TCP in simulation, routing and multicast protocols over the networks in wired and wireless. It has invented in 1989 in a variant of real network simulator. It was supported by DARPA in the year 1995 through the VINT project at USC/ISI, XEROX PARL, UCB and LBL. It is a unix -based. Then the Ns2 environment is a research community. It is an object oriented support in data network components and in the event of discrete scheduler.
6.1. QoS IN NETWORK SIMULATOR:
When a packet is sent from source to destination there are many service requirements are needed. These service requirements are governed by the end-user according to network needs. Network needs guaranteed preferable services to the user in the terms of bandwidth, delay, jitter, probability of packet loss and end to end user.
6.2. ADVANTAGES OF NS-2:
Ns2 is free for algorithms, models, accessory tools and it is fully equipped of protocols. It is a ideal choice for the number of tools and scientific acceptance. Ns2 is a short description architecture and it is a complex software. There is a option for adding new components and to write several software and the data flow in multi step.
6.3. DISADVANTAGES:
In Ns2 it is not quite easy to use new models in BISON, because there are different level and a different scenario. The documentation held in this cannot help at this point. The problem is solved by newsgroup dynamically and by browsing the source code.
7. GLOMOSIM:
GloMoSim is a scalable simulation environment. Actually the execution in utilization for the large communication network is in high-fidelity models. There is an extreme challenging in commercial environment and in the military infrastructure. There is a mixture in the network traffic for the data, voice and imagery. The PARSEC simulation environment are built by the wireless networks in glomosin scalable simulation library. Two mobility models are supported by Glomosim..
8. TESTBED:
Test bed is experiment for large development projects. It is a large environment. In computer security it is a medium-scale repeatable experiment for conducting the experiment in malicious code. The design of the test bed is comprised of hardware and high end pc's for experiment in nodes and the software control. The needs of the test bed users are the software control usability and flexibility in the software. The cost for making the software is more than the test bed hardware. In security issues it is more secure and it has more components to avoid from damaging.
9. CONCLUSION:
Thus the test bed is more expensive than simulation software. In Simulation software the cost is less and it is more secure comparing with test bed. Hereby I conclude that simulation software is better when comparing with test bed.
10. REFERENCES:
