Show simple item record

dc.contributor.advisor Gadallah, Yasser
dc.contributor.author DeifAllah, Dina Salah
dc.date.accessioned 2017-05-14T07:34:05Z
dc.date.available 2018-05-14T22:00:17Z
dc.date.created Spring 2017 en_US
dc.date.issued 2017-05-14
dc.identifier.uri http://dar.aucegypt.edu/handle/10526/5071
dc.description.abstract Wireless Sensor Networks (WSNs) technology is currently considered one of the key technologies for realizing the Internet of Things (IoT). Many of the important WSNs applications are critical in nature such that the failure of the WSN to carry out its required tasks can have serious detrimental effects. Consequently, guaranteeing that the WSN functions satisfactorily during its intended mission time, i.e. the WSN is reliable, is one of the fundamental requirements of the network deployment strategy. Achieving this requirement at a minimum deployment cost is particularly important for critical applications in which deployed SNs are equipped with expensive hardware. However, WSN reliability, defined in the traditional sense, especially in conjunction with minimizing the deployment cost, has not been considered as a deployment requirement in existing WSN deployment algorithms to the best of our knowledge. Addressing this major limitation is the central focus of this dissertation. We define the reliable cost-optimal WSN deployment as the one that has minimum deployment cost with a reliability level that meets or exceeds a minimum level specified by the targeted application. We coin the problem of finding such deployments, for a given set of application-specific parameters, the Minimum-Cost Reliability-Constrained Sensor Node Deployment Problem (MCRC-SDP). To accomplish the aim of the dissertation, we propose a novel WSN reliability metric which adopts a more accurate SN model than the model used in the existing metrics. The proposed reliability metric is used to formulate the MCRC-SDP as a constrained combinatorial optimization problem which we prove to be NP-Complete. Two heuristic WSN deployment optimization algorithms are then developed to find high quality solutions for the MCRC-SDP. Finally, we investigate the practical realization of the techniques that we developed as solutions of the MCRC-SDP. For this purpose, we discuss why existing WSN Topology Control Protocols (TCPs) are not suitable for managing such reliable cost-optimal deployments. Accordingly, we propose a practical TCP that is suitable for managing the sleep/active cycles of the redundant SNs in such deployments. Experimental results suggest that the proposed TCP’s overhead and network Time To Repair (TTR) are relatively low which demonstrates the applicability of our proposed deployment solution in practice. en_US
dc.format.extent 146 p. en_US
dc.format.medium theses en_US
dc.language.iso en en_US
dc.rights Author retains all rights with regard to copyright. en
dc.subject Wireless Sensor Networks en_US
dc.subject Deployment en_US
dc.subject Reliability en_US
dc.subject Stochastic Optimization en_US
dc.subject Genetic Algorithm en_US
dc.subject Ant Colony Optimization en_US
dc.subject Local Search en_US
dc.subject Topology Control Protocol en_US
dc.subject.lcsh Thesis (Ph.D.)--American University in Cairo en_US
dc.title Reliable cost-optimal deployment of wireless sensor networks en_US
dc.type Text en_US
dc.subject.discipline Electronics Engineering en_US
dc.rights.access This item is restricted for 1 year from the date issued en_US
dc.contributor.department American University in Cairo. Dept. of Electronics Engineering en_US
dc.description.irb American University in Cairo Institutional Review Board approval is not necessary for this item, since the research is not concerned with living human beings or bodily tissue samples. en_US
dc.contributor.committeeMember Moustafa, Mohammed
dc.contributor.committeeMember El Ezabi, Ayman


Files in this item

Icon

This item appears in the following Collection(s)

  • Theses and Dissertations [1841]
    This collection includes theses and dissertations authored by American University in Cairo graduate students.

Show simple item record