Research On Topology Control And Performance In Mobile Ad Hoc Networks

The self-organizing Mobile Ad Hoc Networks (MANETs) can be deployed fleetly sinceno infrastructure is required for the network construction. Initially, MANETs are designed toprovide emergency communications for temporary scenarios. As the technology advancing,MANETs will find more and more applications in the future.The dynamic changing topology, which is one of the most perceptible characteristics ofMANETs, has a significant impact on the network performance. To this end, this dissertationfocuses on designing topology control mechanisms to optimize the network performance. Thecontribution of this dissertation includes:1) As an important resource, the energy of nodes determines the network lifetime. Sincethe energy consumption of data transmissions is affected by the transmission paths, wepropose a minimum energy path based topology control algorithm. The algorithm pre-serves both network connectivity and the minimum energy paths between any pair ofnodes with only local neighborhood information, and improves the energy efficiency forend-to-end data transmissions.2) Network capacity is the second main issue to be addressed here since it is the key re-source of wireless networks. This dissertation derives a closed-form equation for the net-work capacity in MANETs, and further studies the relation between network topologyand network capacity. Based on the analytical result, we propose a network capacity-optimized topology control algorithm. The algorithm introduces cooperative communi-cations to improve the capacity and reliability of wireless channels, and configure the linkconnections by selecting the appropriate transmission method and relay node to moder-ate the network interference. Furthermore, it can be applied to MANETs directly, whichrequires topology reconfigurations frequently due to the unreliable wireless channels andnode mobility.3) Network topology determines the efficiency of routing protocols to a large extent sincerouting is based on the network topology. The frequent re-routing operations caused bydynamic topology will not only introduce broadcast storms to the network, but also de-graderoutingefficiency. SincethisproblemwillbemoresevereinCognitiveRadio(CR)- MANETs,thisdissertationproposesaprediction-basedtopologycontrolalgorithm,whichcan construct a reliable network topology by predicting the link available time. It enablesrouting protocols to have cognitive ability for the dynamic the environment to improvethe performance of CR-MANETs.4) An inappropriate configuration of security protocols may consume a large amount ofnetwork resources, particularly in MANETs using cooperative communications. A jointdesign of security protocols and topology control can allocate the network resources ina network-wide perspective and optimize the entire network performance. This disserta-tionanalysesthethroughputofanauthenticationprotocolandenhancestheauthenticationprotocol for MANETs using cooperative communications. Then a topology control basedapproach is proposed to jointly optimize the authentication protocol and network perfor-mance without compromising communication security.