We have two MS theses coming up, one tomorrow and the other the following day.
Dr. Hayes has allowed them to be counted towards your seminar requirement, since there have been fewer seminars this semester due to Covid-19.
Aditi Khare, MS CPE
Title: Malware Detection in Internet of Things Using Opcodes and Machine Learning
Defense date: December 2nd, 2020, 2pm to 3pm
Committee: Dr. Avesta Sasan, Dr. Khaled Khasawneh and Dr. Sai Manoj Pudukotai Dinakarrao
Abstract: In the recent years, the exponential growth of Internet of Things devices has caused a huge security threat. These devices are being deployed even
before being secured. Most of the IoT devices are either unsecured or weakly secured and attackers are taking advantage of this. Even if one IoT device gets infected, it has the potential to spread the malware to the entire network. Obfuscation techniques
like polymorphism are being used by hackers to avoid detection.
Gowtham Tummala, MS ELEN
This research is focused on polymorphic malware detection in Internet of Things networks using opcodes and machine learning. ARM-based malware was used for testing because of the large share of ARM-based IoT platforms making it more
indictive of real-world attacks. Opcodes were extracted by disassembling the dataset using the IDA Pro disassembler. A sequentially ordered dataset of the opcodes was created to be used for detection. Four different datasets namely Dmalware, Dgoodware, Dunseenmalware
and Dunseengoodware were created. A polymorphed version of the unseen malware dataset was also created to test the utility of the approach in polymorphic malware detection. We used the sequential pattern mining algorithm, Mind the Gap: Frequent Sequence Mining,
to mine the most frequent patterns in malware. These Maximal Sequential Patterns aka MSPs were categorized based on their functionality using ARM resources.
Three different approaches were tested and compared. The first approach was to create an opcode-rank dictionary based on opcode frequency in the malware dataset to create vectors for machine learning classification. The second approach
used the frequency of MSPs to vectorize the given dataset while the third approach used the MSP type as a feature for detection. Machine learning classifiers like Decision tree, KNN, Random-Forest, SVM and AdaBoost were used to detect malware as well as polymorphic
malware. It was observed that the sequential pattern mining approaches were faster and more resilient to polymorphed malware. A comparative study showed all classifiers apart from KNN had better performance in the pattern mining approach in detection of polymorphic
Title: Controlled Flight of High DOF Humanoid Robots
Defense date: December 3rd, 2020 at 6:30pm
Committee: Dr. Daniel Lofaro, Dr. Kathleen Wage and Dr. Nathalia Peixoto
Abstract: This work aims to expand the abilities of humanoid robots by implementing flying capabilities to the robot. Humanoid robots are designed to mimic the kinematics of a human, i.e. two arms, two legs, and a head. With this structure, humanoid
robots can be designed and programmed to perform a variety of tasks. Some examples of full-body humanoid robots include Hubo, NAO, iCub (an open-source cognitive humanoid robotic platform), Atlas by Boston dynamics, Honda's ASIMO, Valkyrie from NASA. Some
examples of upper body humanoid robots, i.e. humanoid robots with wheels and tracks instead of legs, include Handle by Boston Dynamics and Mitra by Invento Robotics. Increasing mobility options for humanoid robots make them more versatile.
In this work, we work towards adding the mobility option of flying to the humanoid robot. This was done by adding thrusters to the end-effectors of its high degree of freedom (DOF) robotic arms, we then use control methodology for stabilization.
Specifically, in this work, we study the ability of the latter robot to stabilize over the rotation in the x-axis. To test our algorithms, we built the physical robot via additive and subtractive manufacturing methods. We utilized computer aided design (CAD)
as well as computer-aided machining (CAM) during the manufacturing of the robot.
This resulted in a 6-DOF upper body of a humanoid robot with ducted fans on its end-effectors that operates in the coronal plane. A test fixture that allows for full motion of the robot and ground truth measurement was also created. We used system
identification to create a mathematical model of the dynamics of the robot. This model was then used to design a controller for stabilization over the x-axis. We applied this controller in a simulation then confirmed our results on the physical robot. In the
results, we were able to achieve stability over the x-axis with an overshoot of about30% and a settling time of approximately 5.0 seconds.
Academic Program Coordinator
Department of Electrical and Computer Engineering
George Mason University
4400 University Drive, MSN 1G5
Fairfax, VA 22030