The objective of this project is to secure the emerging wireless networks utilizing cognitive radio and dynamic spectrum access. To avoid harmful interference to primary users, cognitive radios rely on cooperative spectrum sensing to accurately detect idle channels for dynamic spectrum access. This project will develop algorithms to detect and countermeasure the spectrum sensing data falsification (SSDF) attack to cognitive radio networks. The participating student will utilize machine learning techniques to design an effective algorithm to exclude malicious sensing reports.
Intrusion Detection by Deep Learning (Mentor: J. Li, Co-Mentor: C. Xin)
Intrusion detection plays a critical role in protecting modern computer systems. The objective of this project is to conduct intrusion detection by analyzing users' behavior patterns using deep learning techniques. An intrusion can be identified if a user has an abnormal behavior, through the classic machine learning approach that extracts features from user's data. The challenge is that we do not know in advance which features are informative. In this project, we will use deep learning as an automatic feature extractor for intrusion detection.
Public Discourse toward Security Agencies (Mentor: B. Payne, Co-Mentor: J. Still)
The objective of this project is to explore public discourse about federal agencies that protect our computer networks: The National Security Agency (NSA), the Federal Bureau of Investigations (FBI), and the Central Intelligence Agency (CIA). These agencies have as components of their mission the protection of computer networks and the investigation of threats to those networks. There has been public concern over the balance between protecting these networks and individual liberties. Officials must understand these concerns in order to address them effectively. This project will analyze the huge volume of Twitter data that contain variations of the terms NSA, FBI, and CIA.
Human Behavior and Security (Mentor: W. He, Co-Mentor: D. Wittkower)
The objective of this project is to discover the relationship between human behavior and security. As social media, mobile devices, and cloud computing platforms become increasingly prevalent, the research and development of more effective ways to increase internet users security awareness and to encourage them to engage in secure behavior online become critical. The REU student will design customized algorithms to conduct data analysis based on emerging big data analytics and data mining techniques, to find the relationship between human behavior and security.
Ethics for Cybersecurity Professionals (Mentor: D. Wittkower, Co-Mentor: B. Payne)
The objective of this project is to develop new research in ethics for cybersecurity professionals. The student will survey available resources in prescriptive ethics of cybersecurity. The primary area of interest is in the application of fiduciary duties and relationships of special trust to the distinctive moral dilemmas faced by cybersecurity professionals arising between employee loyalty and obligations to the public having to do with privacy and personally identifiable information (PII) management. Following this survey, the student will then conduct further research and write an article, which will afford targeted instruction in prescriptive ethical reasoning on moral dilemmas in data management for cybersecurity professionals.
Face Recognition for Security Applications (Mentor: K. Iftekharuddin, Co-Mentor: Y. Li)
In recent years video surveillance has been widely established in both private and public venues for security. This project plans to use a humanoid robotic platform known as NAO to accomplish complex recognition task such as face and facial expression recognition. The objective of this project is to develop a robust face and facial recognition algorithm using the recently introduced concept in the field of artificial neural network (ANN). The recognition algorithm can then be used in secure applications such as to detect potential intruders as well as identify their facial expression to measure the level of associated threat.