Year: 2018
Type: dissertation
Abstract: Smartphones have become the most preferred computing device for both personal and \nbusiness use. Different applications in smartphones result in different power consumption \npatterns. The fact that every application has been coded to perform different tasks leads \nto the claim that every action onboard (whether software or hardware) will consequently \nhave a trace in the power consumption of the smartphone. When the same sequence of \nsteps is repeated on it, it is observed that the power consumption patterns hold some \ndegree of similarity. A device infected with malware can exhibit increased CPU usage, \nlower speeds, strange behavior such as e-mails or messages being sent automatically and \nwithout the user's knowledge; and programs or malware running intermittently or in cycles \nin the background. This deviation from the expected behavior of the device is termed an \nanomalous behavior and results in a reduction in the similarity of the power consumption. \nThe anomalous behavior could also be due to gradual degradation of the device or change in \nthe execution environment in addition to the presence of malware. The change in similarity \ncan be used to detect the presence of anomalous behavior on smartphones. \n \nThis thesis focuses on the detection of anomalous behavior from the power signatures \nof the smartphone. We have conducted experiments to measure and analyze the power \nconsumption pattern of various smartphone apps. The test bench used for the experiments \nhas a Monsoon Power Meter, which supplies power to the smartphone, and an external \nlaptop collects the power samples from the meter. To emulate the presence of anomalous \nbehavior, we developed an app which runs in the background with varying activity windows. \nBased on our experiments and analysis, we have developed two separate models for reliable \ndetection of anomalous behavior from power signatures of the smartphone. The first model \nis based on Independent Component Analysis (ICA) and the second model is based on a \nSimilarity Matrix developed using an array of low pass filters. These models detect the \npresence of anomalies by comparing the current power consumption pattern of the device \nunder test with that of its normal behavior.
Source: UWSpace (University of Waterloo)
Authors: Soundar Raja James, Robin Joe Prabhahar
Language: English
Subfield: Computer Networks and Communications
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W2785170246