Show simple item record

dc.contributor.advisorKim, Taesoo
dc.contributor.authorHesse, Michael Winfried
dc.date.accessioned2020-09-08T12:45:35Z
dc.date.available2020-09-08T12:45:35Z
dc.date.created2020-08
dc.date.issued2020-05-17
dc.date.submittedAugust 2020
dc.identifier.urihttp://hdl.handle.net/1853/63605
dc.description.abstractIoT devices are known for long-lived hardware and short-lived software support by the vendor, which sets the wrong security incentives for users of expensive IoT systems. In order to mitigate as many known vulnerabilities as possible after the vendor has stopped providing security patches for an IoT device, we present a framework that allows the user to selectively disable single hardware components which provide non-essential features that are associated with said vulnerabilites. In the same way, the framework can also be used proactively to reduce the attack surface of an IoT device by disabling unused features. The user’s selection is enforced by a trusted computing base using different hardware security mechanisms on the ARM platform. To this end, we analyze the common hardware architecture of embedded ARM systems using the example of the Raspberry Pi 4. We conclude that only virtualization provides a fine-grained enough partition capabilities for the purpose of partitioning the hardware into used and unused components. However, we also show how other security mechanisms including IOMMUs and ARM TrustZone could be used as an optimization in some cases. Finally, we give a proof of concept implementation using the Raspberry Pi 4 and the Sense HAT as a simulation of a complex IoT device and show how 6 of its hardware components can be selectively enabled and disabled.
dc.format.mimetypeapplication/pdf
dc.language.isoen_US
dc.publisherGeorgia Institute of Technology
dc.subjectIoT
dc.subjectSecurity
dc.subjectHAL
dc.subjectTCB
dc.titleExtending the lifecycle of IoT devices using selective deactivation
dc.typeText
dc.description.degreeM.S.
dc.contributor.departmentComputer Science
thesis.degree.levelMasters
dc.contributor.committeeMemberSaltaformaggio, Brendan
dc.contributor.committeeMemberAhamad, Mustaque
dc.type.genreThesis
dc.date.updated2020-09-08T12:45:35Z


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record