Hardware Attacks and Mitigation Techniques

dc.contributor.advisorSrivastava, Ankuren_US
dc.contributor.authorBao, Chongxien_US
dc.contributor.departmentElectrical Engineeringen_US
dc.contributor.publisherDigital Repository at the University of Marylanden_US
dc.contributor.publisherUniversity of Maryland (College Park, Md.)en_US
dc.date.accessioned2017-06-22T06:04:21Z
dc.date.available2017-06-22T06:04:21Z
dc.date.issued2017en_US
dc.description.abstractToday, electronic devices have been widely deployed in our daily lives, basic infrastructure such as financial and communication systems, and military systems. Over the past decade, there have been a growing number of threats against them, posing great danger on these systems. Hardware-based countermeasures offer a low-performance overhead for building secure systems. In this work, we investigate what hardware-based attacks are possible against modern computers and electronic devices. We then explore several design and verification techniques to enhance hardware security with primary focus on two areas: hardware Trojans and side-channel attacks. Hardware Trojans are malicious modifications to the original integrated circuits (ICs). Due to the trend of outsourcing designs to foundries overseas, the threat of hardware Trojans is increasing. Researchers have proposed numerous detection methods, which either take place at test-time or monitor the IC for unexpected behavior at run-time. Most of these methods require the possession of a Trojan-free IC, which is hard to obtain. In this work, we propose an innovative way to detect Trojans using reverse-engineering. Our method eliminates the need for a Trojan-free IC. In addition, it avoids the costly and error-prone steps in the reverse-engineering process and achieves significantly good detection accuracy. We also notice that in the current literature, very little effort has been made to design-time strategies that help to make test-time or run-time detection of Trojans easier. To address this issue, we develop techniques that can improve the sensitivity of designs to test-time detection approaches. Experiments show that using our method, we could detect a lot more Trojans with very small power/area overhead and no timing violations. Side-channel attack (SCA) is another form of hardware attack in which the adversary measures some side-channel information such as power, temperature, timing, etc. and deduces some critical information about the underlying system. We first investigate countermeasures for timing SCAs on cache. These attacks have been demonstrated to be able to successfully break many widely-used modern ciphers. Existing hardware countermeasures usually have heavy performance overhead. We innovatively apply 3D integration techniques to solve the problem. We investigate the implication of 3D integration on timing SCAs on cache and propose several countermeasures that utilize 3D integration techniques. Experimental results show that our countermeasures increase system security significantly while still achieving some performance gain over a 2D baseline system. We also investigate the security of Oblivious RAM (ORAM), which is a newly proposed hardware primitive to hide memory access patterns. We demonstrate both through simulations and on FPGA board that timing SCAs can break many ORAM protocols. Some general guidelines in secure ORAM implementations are also provided. We hope that our findings will motivate a new line of research in making ORAMs more secure.en_US
dc.identifierhttps://doi.org/10.13016/M20G40
dc.identifier.urihttp://hdl.handle.net/1903/19399
dc.language.isoenen_US
dc.subject.pqcontrolledComputer engineeringen_US
dc.subject.pquncontrolledhardware securityen_US
dc.subject.pquncontrolledhardware Trojanen_US
dc.subject.pquncontrolledORAMen_US
dc.subject.pquncontrolledreverse-engineeringen_US
dc.subject.pquncontrolledside-channel attacksen_US
dc.titleHardware Attacks and Mitigation Techniquesen_US
dc.typeDissertationen_US

Files

Original bundle
Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
Bao_umd_0117E_17903.pdf
Size:
2.2 MB
Format:
Adobe Portable Document Format