Context- and Uncertainty-aware Computing for Ubiquitous Internet-of-Things-based Applications
In 2003, there were approximately 6.3 billion people living on the planet and 500 million devices connected to the Internet. With the next evolution of Internet, Internet-of-Things (IoT), by 2020 50 billion “things” are expected to be connected to this enormous network. IoT is taking a huge leap in the evolution of the Internet by connecting things, places, sensors, actuators, and other smart mobile devices to enable person-to-object and object-to-object communications. Real-time in-the-field processing of data generated by IoT is crucial for enabling novel ubiquitous applications such as mobile health care, distributed rainfall estimation, real-time food package tracking, in-situ DNA sequencing. IoT applications need to collect and analyze relevant data from an enormous number of (likely wireless) sensing sources to generate accurate and timely results in the field. Therefore, they must be aware of the non-idealities in data and resources as well as of the changes in the context (environment) so to acquire and process only the high quality relevant information.
In this talk, we will introduce a general mobile computing framework to enable futuristic IoT applications as well as a novel cyber interface that utilizes it. Due to the aforementioned reasons, the mobile computing framework is uncertainty (i.e., in data quality, quantity, and resource availability) and context (i.e., location, time activity, and identity) aware. Uncertainty-aware computing capability of the framework handles resource uncertainty (which affects timeliness) and data uncertainty (which affects accuracy). Both context-aware and uncertainty-aware computing improve each other by providing and exploiting the context or uncertainty information. The cyber interface is an interactive tool to describe the IoT application workflow and to connect the physical and cyber domains via the computing framework: such interface will provide middleware to accept input controls using simple diagrams and graphs, to allow users to specify the application accuracy and deadline requirements as well as the sought level of approximation in computing, and to analyze the outputs of the framework.
Dr. Dario Pompili in an associate professor in the Department of Electrical and Computer Engineering (ECE) at Rutgers University, NJ, where since 2007 he is the director of the Cyber-Physical Systems Laboratory (CPS Lab) and the site co-director of the NSF I/UCRC Center for Cloud and Autonomic Computing (CAC). Since 2012, he is also associate director of application collaborations for CPS of the Rutgers Discovery Informatics Institute (RDI2). He received a PhD degree in ECE from the Georgia Institute of Technology in 2007, where he worked in the Broadband Wireless Networking Laboratory. He had previously received a “Laurea” (integrated BS/MS) and Doctorate degrees in Telecommunications and Systems Engineering from the University of Rome “La Sapienza,” Italy, in 2001 and 2004, respectively.
In 2011, Dr. Pompili received the NSF CAREER award to design efficient communication solutions for underwater multimedia applications. In 2012, he received the ONR Young Investigator Program (YIP) award, one out of 26 awarded nationwide, to develop an uncertainty-aware autonomic mobile computing grid framework as well as the DARPA Young Faculty Award (YFA), one out of 51 awarded nationwide, to enable real-time information processing based on compute-intensive models for operational neuroscience. Dr. Pompili published about a hundred refereed scholar publications: with about 4,000 citations, he has an h-index of 21 and an i10-index of 31 (Google Scholar, April'14). Since 2014, he is a Senior Member of both the IEEE Communications Society and the ACM.