Apply…

The deadline has passed

The Program…

The CERIAS Information Security REU Program provides the opportunity for undergraduate students to engage in the forefront of information security research working on individual project areas. Some areas include:

• Assessing Risk of Biomedical Research PHI in of Public HPC Environments [+]

  With the inclusion of the phrase “any other unique identifying number, characteristic, or code” in its description of Protected Health Information (PHI), the Health Information Portability and Accountability Act (HIPAA, 1996) endeavors to protect the privacy of human subjects. However, the ambiguity of this phrase and its possible implications as the state of the art in the life sciences advances raises concerns about the viability of community/shared High Performance Computing (HPC) resources (e.g., the Extreme Science and Engineering Discovery Environment (XSEDE) project) as “safe” environments in which to conduct “in silico” experiments or computationally perform data analyses. In this project, students will have the opportunity to conduct a risk assessment of shared HPC resources that leverages NIST’s “Guide for Conducting Risk Assessments” (NIST SP - 800-30rev1).
• Personal Health Information Security [+]

  This research project will engage students in developing an information security survey instrument using a risk-based approach based on security standards (ISO, COBIT, NIST), legal requirements (HIPAA, FISMA, State privacy laws, etc.), and industry rules (PCI-DSS) to help data stewards understand compliancy requirements while assessing the potential threats to unencrypted data.
• Public Cloud Providers [+]

  The threat associated with public cloud providers varies greatly as compared to a legacy application behind an enterprise boundary. This project will summarize the threat applicable to a public cloud instantiation and compare that to the securitymodel offered by the provider.
• Detecting Patterns of Humor [+]

  According to Raskin’s (1985) hypothesis, all verbal humor is based on the full or partial compatibility of a humorous text with two opposing scripts. Within this general recipe, other components of the joke allow for numerous variations of humorous techniques that individualize the humor authors and deliverers (writers, comedians, etc.). This makes it possible to use humor analysis in order to attribute a text to a certain individual, and important IAS technique. This study is an early step in this direction, so several possible variations of it will be negotiated to fit a student’s own interests and strengths. What we are particularly interested in is what information is not stated explicitly in a joke because it is supposed to be familiar to the hearer/reader and what has to be added as new information. By changing the ratio between what is explicit and what is implicit, one can manipulate the length, verbosity, and style-genre of the joke, thus enriching and manipulating the established patterns in humor, which will contribute to the computational identification of individual humor styles.
• Quantifying Informativeness and Novelty of Text [+]

  The study is a continuation of ongoing work on determining how new and informative a new text is in relation to the ones already processed by the system. It is part of a computational semantic approach to develop computer applications that closely emulate human language and information processing ability. The possibility to develop an Internet search that is much more intelligent that Google because it actually understands the user’s query rather than just searching for the character strings is quite exciting, and this is only one of the numerous ways, many of them cyber security related, that an advanced notion of informativeness can improve in our understanding of and emulating the human mind. We will compare new texts to old ones in order to discover what exactly makes a new text seem informative in relation to some old texts but not others. Our purpose will be to formulate computational rules and regularities that are useful for the task.
• Detecting Script-Underlying Recurring Situations in Text [+]

  An exciting property in human language and information processing activity is our ability to aggregate separate statements into larger chunks of information that frequently recur, are familiar, and become associated with standard routines, such as going to a restaurant or boarding a plane. The current computational implementations of language and information processing do not yet have this capability. We will look at various examples of such chunks of information, often referred to as scripts, and figure out how we, humans, recognize them after seeing just a phrase or two and think about programming the computer to do the same.
• A security-enabled interface for humanoid robot-to-robot communication, within a HARMS Model. [+]

  Inter-robot communication interfaces a number of robots, working in a common task domain. In the case of mobile, autonomous robots the communications medium between the robots is wireless and therefore subject to security compromises. To provide a secure interface between robots a secure layer can be added to a HARMS Model (Humans, software Agents, Robots, Machines, Sensors). In this research project, humanoid robots will be used and a secure module will be added to HARMS networking and communication layers, to enable the robots to work as a team and be insulated from malicious intrusion or dereliction of task, due to external, unsafe influence.
• Deriving Intelligence from Encrypted Stream [+]

  Side channel analysis of a secure communication system is a very interesting technique of gaining valuable information not from the substance of the implemented cryptosystem itself, but rather from its implementation characteristics or its usage specific characteristics. This type of cryptanalysis poses challenges to cryptosystem designers to pay special attention to potential information leakage from cryptosystems’ practical implementations in various communication environments. The significance of the derived intelligence from an encrypted system emphasizes the need for collaborative development of both encryption techniques and conversational communication protocols, in order to adopt security level of the overall communication that will minimize any potential information leakage to the adversaries. This project will evaluate some of the current side channel analysis techniques in order to identify practical security improvement that will be suitable for implementation of the current VoIP networking implementation. Obtained results may be used as input for further communication protocol adoption that will prevent side channel analysis of encrypted voice stream, and leveraging the overall security level of VoIP based communication.

Program Details…

• Duration: 8 weeks (5/28/2013 - 7/20/2013)
• Stipend: $500/weekly
• Housing, meals and travel provided
• Weekly pizza seminar series
• Midterm and final mini-symposium
• Tour the IAS labs on campus
• Social Events

Housing

• Hilltop Apartments
• Hilltop Virtual Tour

Where…

Purdue University, West Lafayette, Indiana

Nondiscrimination Policy Statement

http://www.purdue.edu/purdue/ea_eou_statement.html

This program is sponsored by the National Science Foundation under grant number 1062970.