“Moving from clouds to mobile clouds”: cost-efficient privacy-aware mobile content delivery mechanisms

Download files
Access & Terms of Use
open access
Copyright: Thilakarathna, Dinesh
Altmetric
Abstract
Smart mobile devices are being used to create, consume and share a variety of user data with others, which has led to a significant growth in the use of social networking services and services that support the distribution of user generated content. Service providers are making available more free services, as they are able to monetise user data. As a result, mobile data traffic is growing exponentially and placing heavy demands on mobile networks. This has a two-fold impact on users. Firstly, the communication costs of users are increasing. Secondly, the monetisation activities of service providers pose threats to user privacy and control of their data. Therefore, it has become vital to develop content delivery mechanisms that minimise communication costs, improve user privacy and provide more user control, without breaking the existing digital services eco-system. This thesis introduces three novel mechanisms referred to as User-Stash, MobiTribe and Yalut, which address the three issues: cost, privacy and control of data. It demonstrates the viability of those mechanisms through analytical and experimental evaluations. User-Stash demonstrates that it is possible to reduce mobile network traffic by exploiting the transient co-location of mobile users and spatio-temporal correlation of content popularity. User-Stash selects a set of devices to become crowd-sourced stashes that cache popular content for the benefit of others nearby. MobiTribe enables decentralised social networking services on smart mobile devices exploiting the time elasticity of social networking content and the pervasive availability of high speed low-cost networks such as WiFi, to effectively replicate user data on the devices of their trusted friends whilst minimising the communication costs as well as providing privacy and user control. Yalut augments MobiTribe to provide further cost benefits to users using opportunistic communication where the users are geographically clustered into communities. The novel solutions are based on a set of dynamic time-aware centrality metrics that identify most influential users to propagate content with minimum content delivery delay. In addition to analytical evaluations of the above mechanisms, the thesis demonstrates the practical feasibility of these mechanisms by developing mobile applications that can be used on commodity smart mobile devices.
Persistent link to this record
Link to Publisher Version
Link to Open Access Version
Additional Link
Author(s)
Thilakarathna, Dinesh
Supervisor(s)
Seneviratne, Aruna
Creator(s)
Editor(s)
Translator(s)
Curator(s)
Designer(s)
Arranger(s)
Composer(s)
Recordist(s)
Conference Proceedings Editor(s)
Other Contributor(s)
Corporate/Industry Contributor(s)
Publication Year
2015
Resource Type
Thesis
Degree Type
PhD Doctorate
UNSW Faculty
Files
download public version.pdf 9.59 MB Adobe Portable Document Format
Related dataset(s)