UT Institute of Computer Science Graduation Theses Registry

Name

Oluwaseni Joshua

Abstract

Emerging smartphone technologies has experienced a geometric increase and is currently still on the rise. People use the smartphone for their day-to-day activities such as sending emails, sharing photos and videos through various peer-to-peer social network hubs and so on. In the last few years, the smartphone has experienced massive technological advancements and innovation with respect to its processing capabilities and can now be used to perform complex, resource-intensive tasks in advanced applications like video editing and processing, and object recognition. Although most smartphones have been greatly augmented to handle advanced applications with complex computational needs, they are still limited in terms of their energy resources i.e. battery life. Battery technology has not evolved as rapidly as other areas of the smartphone and so the execution of computational-intensive tasks would cause its rapid depletion; evidenced by the need to constantly charge the device battery. Many techniques have been proffered to maximize energy conservation on mobile devices. Some of which are slowing down the CPU, or shutting off the screen when idle. Among these, the most notable technique for conserving smartphone energy is computation offloading. This basically involves the transfer of the processing of certain tasks from a resource-constrained smartphone to a remote, resource-rich device thereby facilitating energy conservation on the smartphone. This is a fairly large research area and numerous contributions have been made towards advancement in this field. However, much work is yet to be done with regards to energy conservation through offloading during recurrent resource-intensive processing. In this research study we aim to reduce energy consumption during continuous, energy-intensive processing. We consider context-awareness in proposing a scheduling model that could potentially minimize the speedy depletion of mobile device energy thus achieving our aim. We propose a service-oriented framework towards enabling energy-optimal task execution through a task scheduling offload algorithm. We develop a proof-of-concept prototype on an Android device to demonstrate and evaluate the framework’s energy conserving capabilities.

Graduation Thesis language

English

Graduation Thesis type

Master - Software Engineering

Supervisor(s)

Assoc.Prof.Dr. Satish Narayana Srirama, Dr. Chii Chang

Defence year

2015