Mahadevan Gomathisankaran, Ph.D. - US grants

The proposed research targets fundamental work in sensor networks. Specifically the work addresses the central problem that distributed inference systems need to be robust to a wide variety of sensing, and channel impairments. The effort seeks to design a suite of robust, distributed detection and estimation algorithms for sensor networks and Net-Centric applications. A comprehensive approach to robust distributed inference for a wide range of communication channels and topologies, with minimal assumptions about the sensing noise, is proposed.

The outcomes of the proposed work have the potential to greatly impact many Net-Centric applications including environmental, military and health care that require aggregation of data from sensors which are linked only through wireless means. The work is supported by the Industry Advisory Board as well as individual industry members of the center and has the potential to extend the center?s portfolio while potentially attracting new members. The PI has collaborative TUES Phase 3 software project with HBCU colleges that will be tapped on for recruitment of minority students for training in the Ph.D. program associated with this I/UCRC FRP.

Understanding the risk and reliance on specific system security practices is critical in order to formulate a coherent strategy on safe cloud adoption. This work will investigate techniques for security risk evaluation for cloud computing with specific focus on the two areas of off-line risk management and on-line trust evaluation. The research will develop a project assessment model to capture relations between threats, tasks, assets and business logic; develop a vender assessment model to evaluate the vender?s share of security responsibility; and develop a cost benefit trade-off methodology for optimal cloud adoption strategy.
Based on these, will develop an on-line assessment methodology for cloud service providers assessment based on different applications, services and vendor compositions.

With the proliferation of cloud architectures, it is critical that their security be assessable and issues addressable. The proposed work has the potential to enable loud users to gauge the trust-worthiness of individual cloud elements or aggregated cloud resources and to analyze and suggest appropriate risk mitigation strategy. The work is supported by the Industry Advisory Board and will engage multiple industry members of the center. The center will involve graduate students and be integrated in graduate course offerings. Industry tutorials will also be developed and offered.

This workshop on "Airborne Networks and Communications" in conjunction with the annual American Institute of Aeronautics and Astronautics (AIAA) Infotech@Aerospace Conference will be held in Boston, Massachusetts, on Aug. 21, 2013. The workshop will provide a discussion forum for researchers in disciplines important for airborne networking, including wireless communications, autonomous air vehicles, routing planning, cyber-physical security, and test beds, and important to real-world situations such as in disaster relief during which communications, coordination, and sensing are critical. This project provides travel support for invited speakers and for U.S. graduate students.

The project funds the organization of a workshop on "Airborne Networking and Communications". The workshop, colocated with the ACM MobiHoc Symposium, will be held on August 11, 2014, in Philadelphia. The workshop will be the third in a series that started in 2012. Airborne Networking is an emerging area of cyber physical systems - and issues of control, processing, and networking lie at the heart of this field. It is important to build a foundational science to support this area as unmanned vehicles become more prevalent. The workshop brings top researchers in area together to help develop a research agenda and continue to further the field. In addition, strong outreach to undergraduate students for participation is a key component of the NSF funding.

In order to stop the Ebola Epidemic it is essential to do effective "contact tracing." The problem is complex as the contact tracing has to be done retroactively after a patient is diagnosed with the disease. Any lapse in contact tracing could potentially fail to track citizens at risk and spreading of the disease. Ad hoc tracing, relying on the infected carrier?s recollection of places visited and people met may lead to inaccurate tracking. Similarly, relying on the recollections of individuals who may have come into contact with the carrier cannot accurately identify those who are at high risk and those who are at low risk of contracting the disease. This project proposes a framework that uses existing readily available technologies that can do effective, automated contact tracing without compromising the privacy of the users. The framework has two parts: aggregator - which collects information from the users, and analyzer --- which processes the information to do contact tracing. This project will develop advances in algorithms to minimize false positives, storage features for context and social data that is both scalable and will preserve privacy of potential contacts.

The best current method for managing the Ebola epidemic currently is through contact tracing; that is, retrospectively identifying anyone with whom the affected person has come in contact. The current contact tracing system is time intensive and error-prone. The proposed system will do the tracing automatically, using a scalable, privacy preserving method that draws data from the affected person?s social network and smart phone context. The proposed framework is not only applicable to contact tracing for Ebola, but also can be used for contact tracing of other infectious diseases, as well as emergency preparedness in the case of bio-terrorism attacks. The technologies developed as part of this project, such as privacy preserving storage architecture, can also be applied to other problems.