.FLYINGHEAD WIRELESS INFRASTRUCTURE
.TITLE Critical ad-hoc networking features
.AUTHOR Humayun Bakht
.SUMMARY Mobile ad-hoc networks are highly dynamic. In essence, the critical features of mobile ad-hoc networks can broadly be classified as Quantitative and Qualitative feature types. In this article, Contributing Editor Humayun Bakht presents a brief overview of some of the critical features of mobile ad-hoc networks.
.OTHER
Mobile ad-hoc networks, or "short live" networks, operate in the absence of any supporting infrastructure. Therefore, they offer quick and easy network deployment and are ideally suited for situations where either no supporting structure is available or to deploy one isn’t feasible.
Mobile nodes in mobile ad-hoc networks play an important role in establishing communication between various devices in the network. These mobile nodes are free to move and organize themselves in an arbitrary fashion. Each user is free to roam about while in communication with others. The path between each pair of the mobile devices may have multiple links, and the radio between them can be heterogeneous. This organization allows an association of various links to be a part of the same network.
In situations where networks are constructed and destructed in an ad-hoc manner, mobile ad-hoc networking is an excellent choice. The idea of mobile ad-hoc, or packet radio networks, first emerged in the early 1970s when they were first implemented to conduct a feasible report about the performance of radio devices in providing communication in a battle field environment.
However, with the development of notebooks, etc. the idea of commercial ad-hoc networks was developed. In this article, I’ll present a brief overview of the critical features of mobile ad-hoc networks. This article should be helpful to readers who have already gained a fair understanding of these networks through some of my previous articles.
Mobile ad-hoc networks are highly dynamic. Node mobility introduces certain scalability problems in mobile ad-hoc network protocols. When network topology changes frequently, control messages have to be sent between nodes so that new routes are found and propagated throughout the network.
.BREAK_EMAIL To learn more about the critical features of mobile ad-hoc networks, click here.
In such environments, it’s reasonable to expect that when topology changes happen there might be a short period where a lot of control messages will propagate across the network to distribute the new destination paths. Therefore, the protocol designer should make provision for highly dynamic and fast adapting algorithms that minimize control messages, and attempt to utilize long-lived routes to the maximum extent.
Minimising control messages is also essential, because of the additional load they place in the bandwidth-constrained wireless links. There’s at least an order of magnitude difference between wired and wireless data rates, with the former having a standardised 100 megabits/sec in a local Ethernet, and the latter having a nominal bit rate of 10Mbit/sec in the best case. The techniques used to reduce control messages must strike a balance between the minimum amount of messages and flooding the network each time the topology changes by keeping network state information in each node.
The former has the side effect that in an ever changing network topology, stale routes will appear often. While when the latter is applied to a network containing nodes with high rate of mobility, it might result in control messages consuming all the available bandwidth. This could also lead to very slow network convergence where nodes contain either incomplete or out-of-date views of the network topology.
Saving battery power in the participating nodes of a mobile ad-hoc network is an important challenge. In many kinds of mobile ad-hoc networks, mobile nodes usually rely on exhaustible means for providing energy, such as batteries. For these nodes, energy conservation suddenly becomes an important design decision.
Nodes with low battery power may decide to enter a power saving mode when they having nothing to send, or until another high priority event is generated. This behaviour might affect the way the whole network is operating, since each node is responsible for forwarding another node’s packets in addition to its own. If nodes decide to become "selfish" and break the collective and cooperative nature of mobile ad-hoc networking by not forwarding another node’s data, the mobile ad-hoc architecture is endangered. A multitude of other problems and design trade-offs are concerned with power utilization in such networks, and this particular area is becoming the focus of increased attention.
In summary, the critical features of mobile ad-hoc networks could broadly be classified as Quantitative and Qualitative critical feature types. In the following paragraph I’ll list some of the quantitative and qualitative features of mobile ad-hoc networks.
.H1 Quantitative features
The following are some of the quantitative features of a mobile ad-hoc network.
Network settling time can be defined as the time required for a collection of mobile wireless nodes to automatically organize itself and transmit the message reliably. Network Join Time is the time required for an entering node, or groups of nodes, to become integrated into the mobile ad-hoc network. Network Depart time is the time required for the network to recognize the loss of one or more nodes and reorganize itself to route around the departed nodes.
Network Recovery time is the time required for a collapsed portion of the network due to traffic overload, or node failures, to become functional again once the load is reduced or the nodes become operational. Frequency of updates or network overhead can be defined as the number of control packets required in a given period to maintain proper network operation in an ad-hoc network. Memory requirement is the storage space requirements in bytes, including routing tables. Network scalability is the number of nodes the mobile ad-hoc network can scale to and reliably preserve communication.
.H1 Qualitative critical features
Some of the critical qualitative features are the location information of mobile nodes. Does the routing algorithm require local or global knowledge of the network? The effect of topology changes, or how routing algorithms cope with the frequent topology changes in an ad-hoc network is important as well, i.e. does it need complete restructuring or only incremental updates?
Also of interest is the adaptability to different radio communication environments. What mechanisms are adopted to support radio communication at different radio frequency, i.e. do nodes use estimated knowledge of fading, or shadowing of multi-user interference on links in their routing decisions? Power reservations are also important. Does the network employ routing mechanisms that consider the remaining battery life of a node?
Is it single or multi-channel? Does the routing algorithm utilize a separate control channel? In some applications, multi-channel execution may make the network vulnerable to countermeasures. Is it unicast or multicast? Does the routing algorithm perform efficiently on unidirectional links, e.g. if bi-directional links become unidirectional?
Another qualitative feature is network security, such as whether or not the routing and MAC layer policies support the survivability of the network in terms of low probability of detection, or low probability of intercept and security. Also important is the quality of services and the handling of priority messages. Or how efficient the algorithm is in terms of the quality of the services of the network? In other words, does it support priority messaging and reduction of latency for delay-sensitive real-time traffic? Can the network send priority messages/voice even when it’s overloaded with routine traffic levels?
And finally, supporting real-time voice and video services. How efficient is the network in terms of supporting real-time communication, i.e. can the network support simultaneous real-time multicast voice or video while supporting traffic loads associated with situation awareness and other routine services?
.H1 Conclusion
In conclusion, quantitative and qualitative features of mobile ad-hoc networks are helpful in drawing a clear guideline for both the users and researchers in this area about the performance of each separate bit of any ad-hoc networks. These features can best be used for the further development, or possibly in the development of new strategies, for different network controls in an ad-hoc network.
.BIO
