Detection of application used on a mobile device based on network traffic
Loading...
URL
Journal Title
Journal ISSN
Volume Title
Sähkötekniikan korkeakoulu |
Master's thesis
Unless otherwise stated, all rights belong to the author. You may download, display and print this publication for Your own personal use. Commercial use is prohibited.
Authors
Date
2018-06-18
Department
Major/Subject
Communications Engineering
Mcode
ELEC3029
Degree programme
CCIS - Master’s Programme in Computer, Communication and Information Sciences (TS2013)
Language
en
Pages
70
Series
Abstract
Smartphones have become very popular over the past years, thus being owned by almost every individual, the devices also follow their owners throughout the day thus having access to a lot of information about their users. Additionally various companies provide additional services through applications on mobile devices which makes them highly interested in what people do with their mobile devices, as it allows perfection of these services. To collect usage data, on top of having user consent, a company must be able to actually see what is happening on the device. But in regards to growing concern about user privacy, operating systems on mobile devices isolate applications limiting their access to only a small part of information of what is happening on the device. Options like running surveys exist, but are highly dependent on honesty of the people and expensive. To gain the information about running applications network traffic can be utilized as more and more devices are constantly connected to the internet. On the other hand, as well as application isolation, the network traffic is also being more and more protected. This thesis starts with reviewing previous works to give a picture of what kind of information can be extracted from mobile device and it's network traffic and how it can be used. The main aim of this thesis is to implement a system that detects the used applications and their running times by combining mobile network traffic with application launch times and using machine learning. To assess the detection quality and scalability thoroughly, several tests are performed. The implemented detection system shows good potential as it achieves near perfect results in optimal conditions, yet to provide these conditions in every case, a lot of work has to be done still.Description
Supervisor
Asokan, NThesis advisor
Mohtaschemi, MikaelKeywords
feature extraction, internet, market analysis, random forest, smartphones