My Ph.D. thesis mainly focused on
decentralized security solutions to
attest, recover and update IoT
networks. We designed and evaluated a
decentralized scheme that distributes
attestation among devices for systems
that work in swarms. Our approach
assures system resilience to node
compromise/failure while guaranteeing
only devices that execute genuine code
remain part of the group. This
decentralized approach was then
extended to a scalable, efficient and
secure mechanism of recovering a
network of heterogeneous low-end
devices in the presence of
self-propagating malware. We
demonstrated that swarm can be scaled
up to thousands of devices, ensuring
guaranteed update of the entire
network. Finally, we explored and
proposed a data-flow verification
based decentralized swarm attestation
mechanism for embedded devices. This
technique focused on verifying runtime
attacks that corrupt the underlying
application’s security critical data
without hijacking its control flow.
Towards making embedded devices’
attestation complete, we combined both
static and runtime verifications,
assuring that the prover device
executes the correct and unmodified
program (i.e., checking if application
is benign). I am currently focusing on
integrating machine learning (ML) and
Blockchain-based communication and
security solutions into cyber-physical
systems. For example, I am working on
”Fusion of Blockchain and Machine
Learning: A case of Secure Smart
Grid.”, AI-Powered Security for IoT: A
Blockchain Enabled Device Twin
Approach and Property-based
attestation in device swarms: a
machine learning approach. These
approaches aim to optimize, control,
and secure systems such as power
distribution and smart grids. After
implementation, I test them on
real-time hardware, simi- lar to what
I did during my Ph.D using actual
Smart IoT devices. These methods are
valuable for establishing secure
learning-based communications and
creating resilient distributed
systems, addressing power grid control
problems, and cybersecurity
challenges. In summary, these
approaches and the verification
methods mentioned above help guarantee
reliable communication, data privacy,
security, efficient energy
distribution, governance, resiliency,
robustness, and automation for
Autonomous Robotic Systems, Internet
of Things, embedded systems security,
hardware security, Multi-Robot
Systems, Autonomous Unmanned Systems,
Human-Robot Interaction, power grid
control problems, and cybersecurity
applications. I am confident that my
research experience would be a good
fit for the Institute. My expertise
and experience in research activities
in the area of Cyber-Security,
Advanced Scientific Computing, IoT,
Networked Embedded Systems and
Blockchain, publishing research papers
as a lead researcher, Conducting
Lectures and guiding students in
Ethiopia and India, mentoring graduate
students at IIT-Delhi, reviewing
research papers for IIT-Delhi and the
American Journal of Computer Science
and Technology, and interacting with
various researchers and stakeholders,
along with my inherent ability to
identify problems and gaps in existing
technology and collaborate with
stakeholders and leaders to address
them, would be a valuable addition to
any research team. Finally, I am
capable of working in any environment
that employs the following and more:
C, C++, Java, C#, MATLAB, Python,
Perl, HTML, Bash, Rust, UNIX/Linux,
NS3, OMNeT++, MySQL, shell scripting,
socket programming, OpenDSS, PyTorch,
and pandas. Sincerely yours, Samuel
Wedaj Kibret (Ph.D.)