Use this URL to cite or link to this record in EThOS: https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.824258
Title: Intelligent mobile ad hoc network management system
Author: Bait Ali Sulaiman, Majdi Mohammed Said
Awarding Body: Brunel University London
Current Institution: Brunel University
Date of Award: 2016
Availability of Full Text:
Access from EThOS:
Access from Institution:
Abstract:
Currently wireless networks have grown significantly and contributed to several fields of technology especially communication. One of the imperative features of wireless networks is providing access to information without considering the geographical and the topological attributes of a user. One of the most popular wireless network technologies is mobile Ad Hoc network. In order to find an appropriate route between two connected nodes, several routing protocols have been suggested and created. Each routing protocol performs best under specific network conditions, such as under relatively low mobility level and highly dense network size. The main attribute to routing protocol performance degradation is connected to changes of the network context and conditions. Up to date, there is no routing protocol that can maintain its performance at high level under all possible context conditions. In this thesis, the introduction of a management system utilizing artificial intelligence and optimisation techniques to be responsible for predicting MANET routing protocol performance behaviour and selecting the best suited one to adapted to the changes in the network conditions to solve the network performance degradation problem. MANET is modelled with the support of Artificial Intelligent (AI) techniques to help in a better understanding of the network performance under different context scenarios, the use of various packet types, and operating with different routing protocols. Thus the main addition made by this research is the use of different techniques to model our mobile Ad Hoc networks in terms of their behaviour that can be utilised for prediction purposes. An additional contribution is utilisation and comparison of different optimisation techniques based on MANET performance models that can be part of the system for choosing the best of the selected five routing protocols based on the network. The determined parameters for the context that affect the network were average mobility, number of nodes, and packet types. I_MMS manages the selection of a routing mechanism to maintain a stable performance by the network. The selected routing mechanism resulted by a minimum value in delay rate, RA, load, and higher throughput.
Supervisor: Cosmas, J. Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.824258  DOI: Not available
Keywords: Artificial intelligence ; Artificial neural network ; Opnet ; Optimisation ; Wireless routing protocol
Share: