School of Information Technology and Electrical Engineering, The University of Queensland, Australia

Biography: Prof. Neil Bergmann has been the Chair of Embedded Systems in the School of Information Technology and Electrical Engineering at the University of Queensland, Brisbane, Australia since 2001. He has Bachelors degrees in Electrical Engineering, Computer Science, and Arts from the University of Queensland, and a PhD in Computer Science from the University of Edinburgh in 1984. His research interests are in computer systems, especially Reconfigurable Computing and Wireless Sensor Networks. His teaching interests are in machine learning, embedded systems and computer systems. He is a member of IEEE, and a fellow of the Institution of Engineers, Australia.

Speech Title: FPGAs as an Enabling Technology for the Internet of Things

Abstract: Wireless sensor networks are becoming the eyes and ears of the Internet, allowing high temporal and spatial sampling of data from both the natural and the built environment. Wireless networks allow rapid deployment and quick reconfiguration. These benefits often require that such sensor nodes are battery powered, perhaps with some energy harvesting. Usually, such sensors are limited in their resolution and accuracy by their limited energy. "Dumb" sensors simply record and transmit raw transducer data streams for subsequent data analysis by powerful cloud processors. The majority of the energy used by such sensors is in the radio transmission of the raw data. Communications energy can be saved, if the data can be analysed, filtered, and compressed on a "smart" sensor node, and only compressed or summary information sent, but this requires energy-efficient on-node data processing. Ideally such processing can make use of modern machine learning technologies that can learn by example, and respond to changes in the environment, but these also require massive computation. Industrial and mission-critical applications for Internet of Things (IoT) require additional complexity to provide security, redundancy and adaptability. Field Programmable Gate Arrays (FPGAs) are known to be very energy efficient for High Performance Computing in Data Centres. This presentation shows that can also be energy-efficient coprocessors for small embedded systems, such as those found in IoT deployments. There are still research problems that need solutions before FPGAs and IoT can be efficiently combined, such as low power sleep modes with fast wakeup, and better design tools.

Keywords: FPGA, Machine Learning, Wireless Sensor Networks, Internet of Things