Type

Text

Type

Thesis

Advisor

Committee members: Das, Samir, R.; Rahmati, Amir; Polychronakis, Michalis; Nagendra, Vasudevan

Date

2020-05-01

Keywords

Device Activity Detection, Internet of Things, Smart Home, Traffic Signatures

Department

Department of Computer Science

Language

en

Source

This work is sponsored by the Stony Brook University Graduate School in compliance with the requirements for completion of degree.

Identifier

https://hdl.handle.net/11401/79109

Publisher

The Graduate School, Stony Brook University: Stony Brook, NY.

Format

application/pdf

Abstract

The popularity of the Internet of Things (IoT), smart home devices specifically, has been tremendous in the past couple of years. Devices like smart speakers, smart fridge, etc. have integrated seamlessly with the lives of homeowners. The main reasons being their accessibility, cost-effectiveness, and energy-efficiency. These devices by nature, continuously communicate with each other and their servers, to automate the day-to-day activities for their owners and in the process generate huge amounts of network traffic. But at the same time, this underlying automation can often occlude the operation and communication between devices from their owners. This thesis is a step towards understanding the causation of smart-home activities, thereby increasing the visibility and transparency in a smart home. In this thesis, we use the network traffic traces to generate packet-level traffic signatures for device activities (change speaker volume, play music, turn ON lights), which can help identify the triggering devices for an observed activity in a smart home, without causing significant computational overhead or storage constraints. By signature matching, we show how the triggering device can be identified by the users. We present our results on our IoT testbed (WINGS Lab in Stony Brook University) and publicly available datasets. We demonstrate that our approach can identify activities and corresponding sources with good accuracy. | 39 pages

Share

COinS
 
 

To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.