Study On Manets And Other Wireless Models Computer Science Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

MANET's are self-configured networks of portable nodes which are connected through wireless, creating a random topology. MANETS can be created easy and flexible in any environment but with limits on wireless access. The nodes are free to move randomly; therefore the network's wireless topology may be random and may change anytime. Insignificant configuration, quick distribution and lack of a central leading administration make mobile ad hoc networks suitable, but not limited to, emergency conditions like natural disasters, military encounters, medical situations etc. Communication is achieved through routes whose nodes transmit packets. These focus mostly on effectiveness with security transferred to weak adversary models.

Educational Applications: for example on meetings or at several conferences, where all the terminals and access points need to be mobile and where people concentrate with laptops, smartphones or other mobile devices with wireless access in a region that support 802.11 network. The need of participants to move around the place, exchange information and communicate without depending exclusively on a fixed access point, MANET effectively implements all the requirements.

Challenges that need to be addressed when dealing with MANETs

One of the biggest issues faced in MANET networks is the routing of network traffic and standard routing protocols do not meet MANET's needs.

Protocols such as AODC, DSR and ZRP had been proposed to offer solutions but there wasn't any permanent solution.

In addition there is lag of security on the physical layer of the network. The random movement of nodes connected to the network and the continue transformation of network's topology restrict a firewall to be set up on MANET networks. Ways to set up and increase security measures vary from network's area and the distance between nodes. For example if the nodes are too far away from each other the risk of an attack increases or if there are too close to a distance of a physical contact, information's could be spread between them.

Finally there is the challenge of quality of service. QoS is basically needed in real time applications, in which communications required to be reliable definable.

In some process and signals transmitted through the network we could have an Ad-Hoc network, in order to have secure and reliable controls of processes'. QoS challenges need to be met.

A critical review of 4 MANET routing protocols and why wired networks routing protocols are not adequate for MANETs.

DVR (Delay Variance routing)

The initial applications in the proactive plans are based on DVR protocol using the distributed algorithm DBF (Distributed Bellmen-Ford) to calculate the shortest path to the network. Examples of DVR protocols include RIP version 1, version 2 and IGRP. RIP is a dynamic routing protocol for local networks, and therefore classified as an Interior gateway protocol (IGP). Both versions are used today, but considered substandard than other improvement techniques such as OSPF and IS-IS. There is also the RIPng for use in IPv6. DVR is responsible to determine the correct route for packets that need to be forwarded. A route request floods the network by the node, so it can create the required route from the replies received. DSR gives the network self-configuration ability, so a need of a pre-existing network infrastructure and an administrator isn't necessary. Also two main mechanisms working together compose DSR so it can perform the discovery and maintenance of source routes in the Ad-Hoc network. Likewise DSR operates a route preservation scheme. This operation uses the data link layer acknowledgments to discover any lost links, if any is detected a route error control packet is sent to the initiating node and the node will remove the error hop from the host's route cache.

DSDV (Destination-Sequenced Distance-Vector)

DSDV is another routing protocol used in ad-hoc networks which is based on DBF algorithm. The main function of the algorithm is to solve problematic loops in routing. Each node preserves a table that holds information for all accessible destinations, the next node that reaches the destination, number of hops that needs to reach the destination and classification number. The nodes occasionally send this table to all neighbors to preserve the topology, which the network adds to its overhead. Every entry in the routing table is noticed with a classification number given by the destination node. Mobile nodes distinguish old routes from new one by their sequence numbers, thereby avoiding the formation of routing loopsE.G. nodes A, B, C. A package will follow the path ABC (according to the routing table of A) to get from A to C. The line of communication between B and C is "broken". B must forward the message back to A to try to get to C. So A takes the message and wanting to resend to C sends back to B since it doesn't knows about the problem. This creates an endless loop. In DSDV the routing information in cases of complete rejection rarely transmitted between the nodes and often in cases of smaller incremental adjustments.

LSR (Link State Routing)

Main characteristic of these protocols is that each node constructs a map in a graph format showing each node to which other nodes is connected. Thus, each node selects the appropriate next node according to network destination. LSR example protocols are OSPF and IS-IS. OSPF, is perhaps the most widely used for large business

networks and the IS-IS is best known for large service networks. Compared to LSR, the DVR protocols have less computational complexity and cost because the messages require more storage space.

AODC (Ad Hoc On Demand Distance Vector)

AODV routing protocol was developed by NOKIA research centers at University of California, the Santa Barbara and the University of Cincinnati by P. Perkins and S. Das. AODV is capable of both unicast and for multicast routing. The main advantage of this protocol is that it has less delay connection from other approaches. A disadvantage is that intermediate nodes can lead to conflict paths, which confusion might be caused by the sequence numbers between the source and intermediate nodes. Even sending multiple answer packages for a route after taking a single package search path can lead to overflow control. In addition sending "hello" messages from the nodes leads to unnecessary bandwidth consumption.

Most of wired networks depend on symmetric links which are always static. But this is not the case with ad-hoc networks as the nodes are movable and constantly changing their position within networks topology.

Furthermore dynamic topology is also the major problem with ad-hoc routing since the topology is not continual. The mobile node might move or intermediate characteristics might change. In ad-hoc networks, routing tables must somehow discard these changes in topology and routing algorithms have to be adapted. For example in a fixed network routing table update takes place for every 30sec. This update frequency might be very low for ad-hoc networks

Performance of MANETs and the simulation tools that are used to investigate their performance.

Layered architecture is designed based on the assumption that each layer performs correctly. This hypothesis holds in wired networks. In wireless networks, the physical layer and its corresponding link layer are very unstable. Performance of upper layers is highly dependent on the lower layers.

MANETS are vulnerable to many forms of attack without some form of network level or link layer security.

Maintain physical security of the transmission media is hard to perform in MANETS

Existing simulation applications, such as NS2, OPNET Modeler, are based Sequential Discrete Event Simulation architecture, meaning a serial recording of system events in a simulation event queue and their implementation based on the time it happens.

A different application called "MANET Simulator" has been developed for a better simulation. The application is not using the same architecture as previous simulator, the continuous movement of nodes, creates many problems in the proper execution of events. However, based on technical multithreaded processing, which results in each node and each packet can act simultaneously and independently from the rest. This ensures the continuous movement of nodes and the routing of packets. Different simulators had been built to test and evaluate certain process and protocols.

Importance of mobility models in investigating MANETs performance and a main mobility model mostly used.

Mobility models are used to describe the movement pattern of mobile nodes. In order to comprehensively simulate a protocol for an ad hoc network, it's necessary to use a mobility model that can accurately represent mobile nodes which will finally operate in the specified protocol.

One mobility model is Random Way Point Model. In this model nodes are located in an area (usually rectangular) of some size and move into this with a constant velocity which is chosen randomly in the interval. Each node selects the point it wishes to reach and moves that point to the selected speed. When it gets to the point it stops for a random time. Then it selects a new position to which it moves and a new speed, regardless of their previous speed, the previous location and waiting time.

Random Way Point model simulates quite realistic motion of nodes in MANET, however, studies on the model, have identified some problems. As time passes, the model does not reach a steady state in terms of speed. However, the speed is constantly decreasing.