Aldmour_2018.pdf (1.83 MB)
Mobile cloud computing for reducing power consumption and minimising latency
thesis
posted on 2023-08-30, 15:48 authored by Rakan A. AldmourMobile phones can be used for basic computations, whilst in contrast smart phones are used for intensive computations, videos, and social media. Processors have become faster, sensors are multi-functional, whereas battery lifetime has had minimal improvement. In view of rich media which involves huge computational complexity, there are greater challenges for mobile users in terms of battery, storage space and transmission time. Therefore, it is vital to send the intensive tasks from mobile to cloud with minimal delay. In this research, a strategy to either save a file on the mobile or offload it to the cloud has been designed through a comparative analysis of two main factors; offloading delay and the power consumption.
An efficient model of offloading process to estimate the battery consumption and the delay was established. Furthermore, a decision engine was designed to choose the most suitable mobile transmission protocol through a comparative study of each route. Additionally, a new model was proposed to compress the file during the 4G offloading technique, which decreased the file size resulting in less delay, enhanced quality of service and secured data integrity. Further innovations were the use of a secondary server when the first was busy and a back up server to prevent repetitive uploading of files.
The results have shown that the newly developed MECCA (Mobile Energy Cloud Computing Application) decreased the processing time by 30% while offloading the files. The use of the secondary server illustrated that the new features saved the mobile battery and reduced the processing time. Moreover, in Second Upload Round (SUR), the improvements in power usage reduction was noticeable whilst using the 4G connection. Overall, the results offer an insight into the offload possibility for different characteristics. The MECCA application was validated through comparative analysis of the main parameters to the developed analytical model which illustrated a match of 96% between the results.
History
Institution
Anglia Ruskin UniversityFile version
- Accepted version
Language
- eng
Thesis name
- PhD
Thesis type
- Doctoral
Legacy posted date
2018-11-07Legacy creation date
2018-11-07Legacy Faculty/School/Department
Theses from Anglia Ruskin University/Faculty of Science and TechnologyUsage metrics
Categories
No categories selectedLicence
Exports
RefWorks
BibTeX
Ref. manager
Endnote
DataCite
NLM
DC