Autonomous Agents-based Mobile-Cloud Computing
Abstract
The proliferation of cloud computing resources in recent years offers a way for mobile devices with limited resources to achieve computationally intensive tasks in real-time. The mobile-cloud computing paradigm, which involves collaboration of mobile and cloud resources in such tasks, is expected to become increasingly popular in mobile application development. While mobile-cloud computing is promising to overcome the computational limitations of mobile devices, the lack of frameworks compatible with standard technologies makes it harder to adopt dynamic mobile-cloud computing at large. In this dissertation, we present a dynamic code offloading framework for mobile-cloud computing, based on autonomous agents. Our approach does not impose any requirements on the cloud platform other than providing isolated execution containers, and it alleviates the management burden of offloaded code by the mobile platform using autonomous agent-based application partitions. We also investigate the effects of different runtime environment conditions on the performance of mobile-cloud computing, and present a simple and low-overhead dynamic makespan estimation model for computation offloaded to the cloud that can be integrated into mobile agents to enhance them with self-performance evaluation capability. Offloading mobile computation to the cloud entails security risks associated with handing sensitive data and code over to an untrusted platform. Security frameworks for mobile-cloud computing are not very numerous and most of them focus only on privacy, and ignore the very important aspect of integrity. Perfect security is hard to achieve in real-time mobile-cloud computing due to the extra computational overhead introduced by complex security mechanisms. In this dissertation, we propose a dynamic tamper-resistance approach for protecting mobile computation offloaded to the cloud, by augmenting mobile agents with self-protection capability. The tamper-resistance framework achieves very low execution time overhead and is capable of detecting both load-time and runtime modifications to agent code. Lastly, we propose novel applications of mobile-cloud computing for helping context-aware navigation by visually-impaired people. Specifically, we present the results of a feasibility study for using real-time mobile-cloud computing for the task of guiding blind users at pedestrian crossings with no accessible pedestrian signal.
Degree
Ph.D.
Advisors
Bhargava, Purdue University.
Subject Area
Computer science
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.